[Mike Norvak]

DanM, on Dec 14 2010, 05:59 PM, said:

or 9800gtx or gtx 260 or gts250/240

Yeah i'm checking and the price for this product (GeForce GTX 460) is too expensive, near to double. So i think gtx9800 or 250 is the best (just about 165 -175 dollars)

And could i run poly fine with the system specifications pointed before?, u know, maps like those on Eternity ,etc.

[Stabs]

run with the gts250 if you can, its basically a more updated version of the 9800gtx

[Sobek]

I'd also note that you can tell something isn't right because the HUD model doesn't appear to have any lighting applied to it at all...

[VinsaneOne]

Roma Loom, on Dec 13 2010, 12:37 PM, said:

Looks like the texture above the door is outdated. It was polymerized long time ago. I'd gladly doublechecked it but I don't recognize this map...

Actually thanks to The_Commander I've looked at this room in e1l2 and have to say that you don't have polymer HRP or it's very outdated:



Uploaded with ImageShack.us

Mr.Deviance, on Dec 13 2010, 11:49 PM, said:

Not sure if you care about it, but what you have there is either very outdated or not working properly.
But then again, with a geforce 6100, the newest version of polymer wouldn't probably run decent anyway.
But I'm just enforcing what has been said above, to confirm that what you are showing in that screenshot looks very outdated/broken.
There are no bumpmaps and the lighting is too bright.
Roma Loom's screenshot shows how the latest version of polymer looks like.


You need to start playing some duke 3D man http://forums2.duke4...tyle_emoticons/default/laugh.gif
I instantly recognized the door the second I saw it.
I've seen every door and every level of Duke 3D from all possible angles hundreds of times.
I would even recognize areas of duke 3d if they were untextured.

Sobek, on Dec 14 2010, 07:44 PM, said:

I'd also note that you can tell something isn't right because the HUD model doesn't appear to have any lighting applied to it at all...

All I know is it looks real good to me compared to what I've been use to seeing all these years. An this......

......it's not that outdated as you see in the log below.

EDuke32 2.0.0devel 20100311
addsearchpath(): Added I:/DN3D/
OS: Windows XP (5.1.2600) Service Pack 1
Large page support available
Loading opengl32.dll
Loading glu32.dll
Initializing DirectDraw...
Using config file 'eduke32.cfg'.
Scanning for GRP files...
Using group file 'duke3d.grp' as main group file.
Using group file 'autoload/duke3d.grp/polymer_hrp.zip'.

[The Commander]

Your eduke version will soon be a year old, and your HRP is out of date.
Update your shit.

[Mr.Deviance]

The Commander, on Dec 15 2010, 07:44 AM, said:

Your eduke version will soon be a year old, and your HRP is out of date.
Update your shit.

Dude, with a 6100 and with winxp sp1 the latest HRP would lag like shit.
The more you have, the more you want. If he's happy with that old version I think it can only mean good news for him. If that old version looks awesome for him and it also runs decent on his 6100, I think is better than have the latest version lag like shit.
I've tried the latest hrp with polymer on a 6600gt for the fuck of it, and it lags like shit on xp sp3 and win 7 x64. So imagine how a win xp sp1 with a 6100 would run the latest version...
I don't even want to think about it.

[Danukem]

Mr.Deviance, on Dec 15 2010, 07:57 AM, said:

Dude, with a 6100 and with winxp sp1 the latest HRP would lag like shit.
The more you have, the more you want. If he's happy with that old version I think it can only mean good news for him. If that old version looks awesome for him and it also runs decent on his 6100, I think is better than have the latest version lag like shit.
I've tried the latest hrp with polymer on a 6600gt for the fuck of it, and it lags like shit on xp sp3 and win 7 x64. So imagine how a win xp sp1 with a 6100 would run the latest version...
I don't even want to think about it.

He should at least get the latest eduke32.exe, though, because it will actually run better than the one from a year ago. Also, Polymer with no HRP doesn't look half bad.

[Mike Norvak]

VinsaneOne, on Dec 15 2010, 02:33 AM, said:

All I know is it looks real good to me compared to what I've been use to seeing all these years. An this......

Try the latest synthesis though : Eduke32 - 1734

[Sobek]

VinsaneOne, on Dec 15 2010, 07:03 PM, said:

All I know is it looks real good to me compared to what I've been use to seeing all these years. An this......

......it's not that outdated as you see in the log below.

EDuke32 2.0.0devel 20100311

That's OK, but just as a note, around August the new builds came into effect that supported HUD lighting and, for a lot of people, some pretty impressive performance improvements.

Definitely update to the newest release in Norvak's post above!

[Stabs]

on the bright side, you will be able to run DNF vinsane http://forums2.duke4...tyle_emoticons/default/smile.gif

[Mr.Deviance]

DanM, on Dec 15 2010, 06:53 PM, said:

on the bright side, you will be able to run DNF vinsane http://forums2.duke4...tyle_emoticons/default/smile.gif

Run DNF with and integrated geforce 6100 on winxp sp1? Are you on ethnobotanical supplements mate?
If the game will come out as a multiplatform game, they will most likely optimize it for multicore cpu's and make it require at least an 8800 gt and 2gb of ddr2 as it's absolute minimum requirements.
Trust me, when I tell you that nothing in this world will make a 6100 run DNF on winxp sp1.
My guess is that it won't even start due to old software to begin with and if by any chance the the os and other drivers get updated, the game will most likely error at launch due to missing shading or other rendering capabilities of that 6100.
If he has winxp sp1 and one year old eduke stuff, he most likely has outdated directx an drivers too.
That pc needs some serious update and upgrading, if he wants to even see one pixel of dnf on his screen.
And I'm saying this with absolutely no exaggeration or irony.

[The Commander]

You do go on a bit to much don't you think Mr.Deviance.

[Mr.Deviance]

The Commander, on Dec 15 2010, 09:37 PM, said:

You do go on a bit to much don't you think Mr.Deviance.

I don't.

[Stabs]

you can nerf the cunt out of any game

i played through unreal and ff7 back in the day on a p120 16mb of ram with no gfx acceleration it wasnt the best but it could be done and bioshit had some patch to disable shaders and let you run it on older cards and thats running on the unreal engine

some peoples definition of running fine is very different to a lot of people, personally iam a constant 60 fps man but some find 30 fine or even 15 fine

[Mike Norvak]

I hav this problem:

Eduke only works with Autoload box on

Any suggestion?

[Mike Norvak]

I made an autoload folder and delete all the textures and models ,etc. to run eduke without the HRP , is the only way i got to make it work so far, really extrange i think...

[Danukem]

Norvak, on Dec 17 2010, 07:19 PM, said:

I made an autoload folder and delete all the textures and models ,etc. to run eduke without the HRP , is the only way i got to make it work so far, really extrange i think...

That is really strange. I wonder why no one else is having the problem.

Did you try deleting your cfg files?

[Mike Norvak]

DeeperThought, on Dec 17 2010, 10:28 PM, said:

Did you try deleting your cfg files?

Yeah, but it´s the same.

Anyways i fixed the problem. The logic solution was to reinstall everything and it works, however i wonder (like u) why the f..ck it happened...

BTW haha i wrote extrange instead of strange, a cognate in spanish : "extrano" http://forums2.duke4...tyle_emoticons/default/unsure.gif

[Hendricks266]

r1736 does not want to build under my MinGW/gcc 4.3.3 setup, and I have a hunch that the problem is on my end because synthesis built just fine.

Build started using "cc -fomit-frame-pointer -funswitch-loops -O2 -fno-stack-protector -W -Wall -Wimplicit -Werror-implicit-function-declaration -funsigned-char -fno-strict-aliasing -DNO_GCC_BUILTINS -Isource -Ibuild/include -Isource/jmact -Isource/jaudiolib/include -Isource/enet/include -D_FORTIFY_SOURCE=2 -fjump-tables -march=pentium3 -mtune=generic -mmmx  -fno-pic -DUNDERSCORES -I../sdk/dx/include  -DHAVE_INTTYPES -DRENDERTYPEWIN=1 -DSUPERBUILD -DPOLYMOST -DUSE_OPENGL -DPOLYMER -DNEDMALLOC"
/bin/sh: /c/cygwin/bin/cc: Bad file number
Failed building obj_win/game.o from source/game.c!
make: *** [obj_win/game.o] Error 1

Perhaps it's time to upgrade to 4.5.1.

[Plagman]

helixhorned, what's the use case for the Makefile changes you made? How do you pass your semantic analyzer as CC? Passing make variables through the commandline should already override definitions, even if they use = instead of ?=. For example, the synthesis script uses the following commandline and it works just fine, even before your Makefile change.

make PLATFORM=WINDOWS CC='wine gcc' CXX='wine g++' AS='wine nasm' RC='wine windres' STRIP='wine strip' AR='wine ar' RANLIB='wine ranlib' PRETTY_OUTPUT=0 NEDMALLOC=0

[VinsaneOne]

Sobek, on Dec 15 2010, 07:50 PM, said:

That's OK, but just as a note, around August the new builds came into effect that supported HUD lighting and, for a lot of people, some pretty impressive performance improvements.

Definitely update to the newest release in Norvak's post above!

I will take that advice.

DanM, on Dec 15 2010, 09:53 PM, said:

on the bright side, you will be able to run DNF vinsane http://forums2.duke4...tyle_emoticons/default/smile.gif

I hope so!

Mr.Deviance, on Dec 16 2010, 12:29 AM, said:

Run DNF with and integrated geforce 6100 on winxp sp1? Are you on ethnobotanical supplements mate?
If the game will come out as a multiplatform game, they will most likely optimize it for multicore cpu's and make it require at least an 8800 gt and 2gb of ddr2 as it's absolute minimum requirements.
Trust me, when I tell you that nothing in this world will make a 6100 run DNF on winxp sp1.
My guess is that it won't even start due to old software to begin with and if by any chance the the os and other drivers get updated, the game will most likely error at launch due to missing shading or other rendering capabilities of that 6100.
If he has winxp sp1 and one year old eduke stuff, he most likely has outdated directx an drivers too.
That pc needs some serious update and upgrading, if he wants to even see one pixel of dnf on his screen.
And I'm saying this with absolutely no exaggeration or irony.

It is a Gateway GT5228 w/Athlon 64 x2 Dual core 4200+ at 2.21 GHz w/1.87 GB Ram. I just installed Win Xp Pro on a fresh, new WD Caviar SATA 320 GB HD, so it still has SP1 untill I find out why I'm having some errors updating to SP2. BSOD is one problem after installing SP2! Had to do a system restore from before that update to get it to boot to my desktop. Damn Microshaft! I have all the latest graphic drivers as well as DX9c. Also have the latest device drivers for all my hardware, but think I may need to update the bios as a last resort if another SP2 install craps out again.

[Hendricks266]

I will no longer be uploading my EDuke32 builds.

[Helixhorned]

Plagman, on Dec 18 2010, 08:54 AM, said:

helixhorned, what's the use case for the Makefile changes you made? How do you pass your semantic analyzer as CC? Passing make variables through the commandline should already override definitions, even if they use = instead of ?=. For example, the synthesis script uses the following commandline and it works just fine, even before your Makefile change.

make PLATFORM=WINDOWS CC='wine gcc' CXX='wine g++' AS='wine nasm' RC='wine windres' STRIP='wine strip' AR='wine ar' RANLIB='wine ranlib' PRETTY_OUTPUT=0 NEDMALLOC=0

For me, putting the variables before the make command works more reliably. Specifically, when I run scan-build like this:

CC=clang RELEASE=0 scan-build make

it will analyze the code as described on their page, while using gcc will compile but not analyze, and putting the variables as arguments to make won't build.

Hendricks: it seems that your default C compiler (the CC variable) is set to Cygwin's, therefore the new Makefile tries to use it and fails.

I'll commit a fix shortly but IMO the user-dependent stuff needs to be ripped out from the 4 Makefiles into a single one.

[LeoD]

Helixhorned, on Dec 19 2010, 05:28 PM, said:

For me, putting the variables before the make command works more reliably. Specifically, when I run scan-build like this:

CC=clang RELEASE=0 scan-build make

it will analyze the code as described on their page, while using gcc will compile but not analyze, and putting the variables as arguments to make won't build.

Hendricks: it seems that your default C compiler (the CC variable) is set to Cygwin's, therefore the new Makefile tries to use it and fails.

I'll commit a fix shortly but IMO the user-dependent stuff needs to be ripped out from the 4 Makefiles into a single one.

Putting "export CC=gcc" (without the quotes) into .bashrc fixes the issue for the time being.

[Mike Norvak]

I didn´t find the correct topic to post this question. Well, i want to change the palete of one single texture in the whole map, it´s an skybox. Some script maybe?

[djdori11]

eduke r1742 is awesome but plz plz try to improve ati cards performence and fix ati cards crashes (with polymer hrp)! after that the game gonna be perfect! thx!

[Helixhorned]

Norvak, on Dec 20 2010, 04:31 AM, said:

I didn´t find the correct topic to post this question. Well, i want to change the palete of one single texture in the whole map, it´s an skybox. Some script maybe?

Sure, nothing easier than that.

for i allsectors, { ife sector[i].floorpicnum XXX, set sector[i].floorpicnum YYY; ife sector[i].ceilingpicnum XXX set sector[i].ceilingpicnum YYY }

I think that this should go into the Mapster32 problems and bugs thread, even though it's not a bug.

[DavoX]

I didn't get half of that http://forums2.duke4...tyle_emoticons/default/tongue.gif

[Mike Norvak]

DavoX, on Dec 20 2010, 12:25 PM, said:

I didn't get half of that

http://forums2.duke4...tyle_emoticons/default/blink.gif Neither i do. I think i shouldn´t ask things as if i knew what i´m talking about. First of all i´ll investigate how a srcript works.

[Helixhorned]

Oh, I see you said "palette", in which case my code does the wrong thing. But conceptually it's super-easy.

for i allsectors	// iterate over all sectors, its index will be i in each iteration
{	// a brace; opens a block
	ife sector[i].ceilingpicnum XXX	// if the tile number of sector i's ceiling *e*quals XXX, ...
		set sector[i].ceilingpal YYY	// set sector i's palette number to YYY; no brace needed because only 1 statement
	ife sector[i].floorpicnum XXX	// same thing with floor
		set sector[i].floorpal YYY
}