[DarkyDan]

Compiled latest polymer build of eDuke on my eeePC, but of course intel integrated graphics don't support such effects so it just crashes without a whimper.

Can't run anything OpenGL on my Windows 7 64bit machine at the moment, ATI drivers don't seem to support it yet? regardless its a rather huge pile of rancid pork if you ask me. I'd love to test this new renderer out! To do so would mean installing Vista -shudders- or even worse.. XP -dies-, or I guess if I was THAT desperate Ubuntu 9.04?

Anyway, keep up the good work gentlemen.

P.S. people on other forums who say 'if I want to play Duke3D I want shiggy old Duke3D graphics' are tools, the gameplay is basically identical with the HRP/Polymost, save for the LACK of severe headaches associated with Y axis mousemove in the original!

[MusicallyInspired]

Heh I never ever used the mouse in 8-bit mode. It's all keyboard, baby. However, with proper 3d looking in Polymost I use the mouse.

[Jimmy]

DarkyDan, on Jun 3 2009, 06:20 PM, said:

P.S. people on other forums who say 'if I want to play Duke3D I want shiggy old Duke3D graphics' are tools, the gameplay is basically identical with the HRP/Polymost, save for the LACK of severe headaches associated with Y axis mousemove in the original!

Someone sounds butthurt.

[WedgeBob]

Well, I have a good enough rig, and I may try this out, providing it won't crash my rig to a million pieces and BSoD everytime I try mapping in Mapster/Build like HRP did.
I assume that there is 64-bit support for Vista this time. I believe that was the trouble that HRP had with me before... Trying to map around the 64-bit environment and a Phenom 9550 CPU/9800GTX+ card...heh.
Better hope that with this new Polymer technology, we'll see the support that I may need to create something awesome. I'll have to recreate some old maps I tried doing back in the '90s.

[supergoofy]

I tried the latest SVN build (Jun 6, 2009) but it crashes after it loads the textures in cache (SIGSEGV crash). The source was compiled with no errors or warnings. The crash happen with Polymost. I haven't tested Polymer.

SVN build of (May 31, 2009) is the most stable of the SVN builds.

You have to download the source of the build released on May 31, 2009 and compile it.

As you see I don't even dare to refer to build numbers, because I don't want to be killed by TX or Plagman.

[inquisitor]

When polymer goes public I'll buy anything needed to run it smoothl a new videocard a new pc anything!
go on like that you rock!

[supergoofy]

Ok, I tested also the newest source as of Jun 7, 2009. The Polymer renderer seems to be faster and working good, but Polymost renderer is broken and gives SIGSEGV crash.

Polymer is still heavy though, even for my system.

[The Commander]

I can also confirm the Polymost crash.

[WedgeBob]

I dunno, maybe it's best I revert back to JonoF's latest port up until Polymer and eDuke32 w/ HRP becomes more stable. Probably the safest move for this moment.

[Jimmy]

The current official release of Eduke32 is stable. The Polymer builds are unofficial development bits, that most of these people weren't supposed to have.

[supergoofy]

Eduke32 is released under GNU license. Thus anyone should have access and can download the source.

People that have some experience on how to compile the source, they can use the compiled binaries for testing. But in respect of TX and Plagman and the other EDuke32 developers, you should never publish these binaries.

[corpy]

supergoofy, on Jun 7 2009, 10:10 AM, said:

Eduke32 is released under GNU license. Thus anyone should have access and can download the source.

People that have some experience on how to compile the source, they can use the compiled binaries for testing. But in respect of TX and Plagman and the other EDuke32 developers, you should never publish these binaries.

So You Say I can compile ver 2.0 Binaries Myself but I Can't give those anyone?
If i'ts so then I try compile...
sry bad english lol...
BTW: eDuke32.0 Is awesome I Hope you guy's relase it Fast !

[supergoofy]

I said that because we respect the very hard work that TX, Plagman, and the other devs put on EDuke32, we should not publish the personally compiled binaries.

Though, you can report bugs here and help TX and Plagman to improve EDuke32.

[TerminX]

WedgeBob, on Jun 7 2009, 10:53 AM, said:

I dunno, maybe it's best I revert back to JonoF's latest port up until Polymer and eDuke32 w/ HRP becomes more stable. Probably the safest move for this moment.

Hi. I've noticed from years on 3DR that you are really, really stupid. Please don't continue that tradition here on Duke4. Thanks!

[engineer]

I was surprised that I was never told that when I started posting here.

[supergoofy]

Report : Latest SVN build Jun 8, 2009 : Polymost is working again. I will test further both Polymost and Polymer.

From some tests that I made, it seems that both Polymost and Polymer are working now.

[edit]
I've done more tests, the last SVN build Jun 8, 2009 seem that it works fine.

[supergoofy]

Report: Latest SVN build Jun 9, 2009 has a crash bug

Run eduke32.exe - On startup window click cancel - Crash error :

AppName: eduke32.exe AppVer: 1.9.9.9 ModName: eduke32.exe
ModVer: 1.9.9.9 Offset: 00128b30


[edit]
newest SVN build fixed it



QUESTION: Can I refer to build numbers when reporting a bug? And how would you like me to report bugs: through posts or p.m. ?

[supergoofy]

I don't know if it's because of the new memory allocator or other optimizations you did, but now polymer is 50% faster on my system. E1L1 is playable now (with polymer hrp). I cannot believe it.

TX and Plagman you are awesome! Thanks.



The new memory allocator takes advantage of dual, quad and generally multicore systems.

[Chip]

supergoofy, on Jun 9 2009, 01:55 PM, said:

The new memory allocator takes advantage of dual, quad and generally multicore systems.

And I once said as a joke that after polymer was done, the next big thing would be to get it multi threaded. But by the looks of it, that comes with polymer!

OK then, new suggestion - once this is finished and released, could you guys think about model tagging? That thing which allows models to attach to other models (someone said that MD3's support this but Eduke32 does not take advantage of it)

[Tea Monster]

Cool. I had to go over to a freinds house to download the compling tools as my 10kbps connection sucks ass. I'll have to build a new version and see how it runs!

I don't know how much a new model format would help the model in the hands problem. I tried to make the duke model so that it could have several animations that would allow it to hold all the different weapons. You'd need basically at least THREE (probably more) different yet complete animation cycles for the model so that it could hold all the different weapons. Then, if someone made a new model of the Shrinker, we would have to animate it bobbing around in space so that the two would match up.

There is no code in EDuke to support matching the player model with a different weapon model. Utilizing the 'sticky verts' option may help this, but I don't know if you would have to cut the arms off the player model so that the hands can move to stick to the weapons.

I think it would be a good idea to work this out, if we are going to do it, before the new wave of models comes out for Polymer.

[vtastek]

I managed to build the source code. I am playing it and enjoying the dynamic lights. There are some stutters at the beginning of levels and some lights(light of the light) can be seen behind walls and noticed a poster flickering but other than those minor issues it is working great. I played for hours.

I didn't see any real-time shadows yet. I will check the ini and console commands now. Scratch that.

I have to tell you it is working great. My fps swings from 400 to 200 all the time. I think my average fps is 250 without shadows. I only saw 60 fps on this occasion:
Fire

If I can't get shadows I will fill a report. Scratch that too.

Edit: I am using the hrp from public svn. Map hacks aren't included there as an update. I guess that is scheduled for later. Scratch. I found the mhk files. While packing, in excitement it looks like I somehow deleted them.

I did grabbed the files from Parkar's post. Will install right now to see some shadows. Svn is the more updated one I think, no?

[WedgeBob]

TX, on Jun 7 2009, 03:31 PM, said:

Hi. I've noticed from years on 3DR that you are really, really stupid. Please don't continue that tradition here on Duke4. Thanks!

What makes you think I would do that?! I'm bringing myself back to the glory days of the '90s again, back to when I was actually GOOD with Duke modding/mapping. I ain't gonna be what I was over the last five years.

[atlas]

for those of us who have not compiled the newest build and are still anxiously awaiting the public release.. can we get some more PR Shots?

[Yatta]

I should post some shots with adjusted ambient lighting to disprove the impression of some people that the new engine is dark.

[Piano Man]

Yatta, on Jun 13 2009, 10:47 AM, said:

I should post some shots with adjusted ambient lighting to disprove the impression of some people that the new engine is dark.

Erm... that would be nice...

(Wipes the drool from his mouth)... would love some new pictures.

[WedgeBob]

I stand corrected, EDuke32 is very stable with HRP with the latest versions... It was Mapster/Build that was unstable, since it BSODs every time I switch from 2D to 3D building (unless that problem's since been corrected without any prior knowledge). JonoF's version of Build didn't do this. I'm hoping that mapping will be more stable now. I'm aware that Polymer isn't available to the public yet, but I'm just hoping that it'll be put to good use sooner than later.

[Benichon]

Some screenshots of mine (with rev1429):
I'm sorry in advance because some texture doesn't have normal texture and i didn't know how to put dynamic light.
Normal textures are done with CrazyBump but i'm newbie so i don't know how to make a good one.

Gallery :
http://img15.imagesh...?g=duke0014.jpg

Other Screen :

[supergoofy]

I believe that r1429 is stable enough to be candidate for official snapshot

[Danukem]

supergoofy, on Jun 13 2009, 12:54 PM, said:

I believe that r1429 is stable enough to be candidate for official snapshot

Being stable isn't the same as being finished, though.

[supergoofy]

that's true

but all the official snapshots are beta builds that considered stable, but not final.