[Richard Shead]

The Commander, on 28 November 2014 - 09:37 AM, said:

Why does the player look 100 meters tall in the video?

[PokeThePower]

MetHy, on 27 November 2014 - 09:04 AM, said:

...a lot was done with Build and RoR after DN3D and after Shadow Warrior.

As far as I know it was almost entirely Alan on his own experimenting, and yes, it would have been for what eventually led to DNF. I don't know if it would have ever wound up on an archive disk and only a slim chance it survived in a dark corner of his hard drives up until the end.

MetHy, on 27 November 2014 - 09:04 AM, said:

Even if DNF's "official" first Build was Quake engine, that was probably for "the sequel to DN3D which wasn't quite called DNF yet", which means that Frederik might not even have the right to show that (if he even wanted to and if it even existed and still existed).

My entire time is post genuine DNF development start and they have one or more of my drives. I don't know if it's an early one or the last one before my "departure" though, but I've been curious too what sort of things might be releasable from that time. There are some really interesting behind the scenes documentations of DNF development but I feel like they aren't supposed to tread those waters.

[Hendricks266]

Wieder, on 28 November 2014 - 03:54 PM, said:

As far as I know it was almost entirely Alan on his own experimenting, and yes, it would have been for what eventually led to DNF. I don't know if it would have ever wound up on an archive disk and only a slim chance it survived in a dark corner of his hard drives up until the end.

There's some stuff on Dirk Jones' drive that warrants further investigation. Of course, TX and I would welcome being sent data from more drives.

Or me going to Denmark and archiving everything professionally.

[Frederik Schreiber]

Wieder, on 28 November 2014 - 03:54 PM, said:

As far as I know it was almost entirely Alan on his own experimenting, and yes, it would have been for what eventually led to DNF. I don't know if it would have ever wound up on an archive disk and only a slim chance it survived in a dark corner of his hard drives up until the end.


My entire time is post genuine DNF development start and they have one or more of my drives. I don't know if it's an early one or the last one before my "departure" though, but I've been curious too what sort of things might be releasable from that time. There are some really interesting behind the scenes documentations of DNF development but I feel like they aren't supposed to tread those waters.

The First Build is the Sidescroller. The one afterwards is an old Quake 1 build, which is primarily using Quake 1 assets, and a few new Duke assets, before they switched to Quake 2. Purely for experimenting with BSP based leveldesign, and getting to know the Quake engine.
The switch from Quake 1 - 2 literally happened within months, so there isn't a lot of content in the Quake 1 builds.
Quake 2 also lasted briefly, and everything you see in the 1998 trailer is literally 90% of what was done. The announcement to switch to Unreal Engine, actually happened before E3 1998.
The first Unreal builds, also consisted primarily of Unreal 1 assets. In 99 they had early versions of a few levels, and weapons done, which resulted in the 99 screenshots. Afterwards development sped up, and in late 2002, the "DNF 2001" build was at it's best, before changing to a new dynamic lighting renderer, which focused on normal maps instead (Doom 3 style).
From then on, the game completely changed, and almost everything was scrapped in small segments. All levels turned out black, as a result of implementing a fully dynamic lighting engine (Just like Doom 3, and Deus Ex 2). Assets were added, replaced, and ultimately large parts of the game was completely scrapped and redone.
The 2002 - 2009 builds ultimately makes up the version of the game you guys know today.

[X-Vector]

That confirms my earlier suspicion that DNF was FUBAR'd some time around 2002/2003.

[icecoldduke]

Frederik Schreiber, on 01 December 2014 - 02:57 AM, said:

The First Build is the Sidescroller. The one afterwards is an old Quake 1 build, which is primarily using Quake 1 assets, and a few new Duke assets, before they switched to Quake 2. Purely for experimenting with BSP based leveldesign, and getting to know the Quake engine.
The switch from Quake 1 - 2 literally happened within months, so there isn't a lot of content in the Quake 1 builds.
Quake 2 also lasted briefly, and everything you see in the 1998 trailer is literally 90% of what was done. The announcement to switch to Unreal Engine, actually happened before E3 1998.
The first Unreal builds, also consisted primarily of Unreal 1 assets. In 99 they had early versions of a few levels, and weapons done, which resulted in the 99 screenshots. Afterwards development sped up, and in late 2002, the "DNF 2001" build was at it's best, before changing to a new dynamic lighting renderer, which focused on normal maps instead (Doom 3 style).
From then on, the game completely changed, and almost everything was scrapped in small segments. All levels turned out black, as a result of implementing a fully dynamic lighting engine (Just like Doom 3, and Deus Ex 2). Assets were added, replaced, and ultimately large parts of the game was completely scrapped and redone.
The 2002 - 2009 builds ultimately makes up the version of the game you guys know today.

Do you know why the decision was made to switch from Quake 2 to Unreal 1?

[Richard Shead]

icecoldduke, on 01 December 2014 - 09:02 AM, said:

Do you know why the decision was made to switch from Quake 2 to Unreal 1?

[MYHOUSE.MAP]

It sounds like to me that Broussard tried to play catch up with Carmack and it completely fucked the project over. It reminds me of what happened to Deus Ex: Invisible War, but on a much longer and larger scale (facny dynamic lighting system detroys the game) It's compeltely irresponsible managament to be making a dramatic change like that so far into development. If he hadn't tried to beat Carmack, then I'm guessing that DNF would've been released around 2004ish.

The ultimate irony is that the facny renderer actually makes the game look kinda shit, so the entire effort was a waste of time.

Based on this information, I feel that 3DR announced the game way too early. It should've been announced around the time they settled on the Unreal Engine, not when the team was fucking around and experimenting with Q1's engine and were not intending to keep it long-term.

[High Treason]

I think the odds were somewhat against the announcement though to be fair, because Duke 3D was from a turning point in the industry. Go back a few months before that game and you could still make a game in less than a year/two years with a team of less than 20 people and be up to speed with the competition so announcing your next game in your current one was just the done thing. A few months later, you suddenly needed more time, more people and more money to get things done and this tradition disappeared.

Technology moved very quickly too; computers in 1996 were topping out with the Pentium 200 (Jun 96) and rarely had 3D accelerators, by 1998 this had more than doubled to 450MHz (Aug 98 - though rare) as had the power of graphics accelerators due to nVidia starting to make decent stuff, ATi pushing the envelope with the Rage series and 3DFX doing whatever they did; mostly waste electricity. By 1999 the world had reached around 750-800Mhz and would quickly accelerate past the 1GHz barrier in 2000. This is only taking into account clock frequencies in computers of the era, it is leaving out other innovations, APIs and whatnot. Imagine trying to keep up with that and with the reputation of your previous game to live up to... It'd drive you nuts.

Though I agree, there were very poor management decisions made and it wrecked the game.

[Mr. Tibbs]

Frederik Schreiber, on 01 December 2014 - 02:57 AM, said:

Snip

Thanks for the details.

Matt T. Wood left 3DR in 2003 for Valve. When DNF shipped he said there were quite a bit of his original work in the 2011 shipping game.

Quote

This is so surreal. I just found a set of railing models that I made in the Hoover Dam portion of DNF (which incidentally, is pretty fun).

Here's a feature in DNF that I worked on with 2 other guys back in 2000 or 2001. Cool to see it in the final game.

https://twitter.com/...044144896946176
https://twitter.com/...136928644698112

[icecoldduke]

Mr. Tibbs, on 01 December 2014 - 11:15 AM, said:

Thanks for the details.

Matt T. Wood left 3DR in 2003 for Valve. When DNF shipped he said there were quite a bit of his original work in the 2011 shipping game.


https://twitter.com/...044144896946176
https://twitter.com/...136928644698112

Says the youtube account associated with the video has been terminated :/.

Quote

It sounds like to me that Broussard tried to play catch up with Carmack and it completely fucked the project over.

It's just sad they didn't focus on the fun and focused solely on the tech, which always ends in disaster.

[Mr. Tibbs]

icecoldduke, on 01 December 2014 - 11:30 AM, said:

Says the youtube account associated with the video has been terminated :/.

It's just sad they didn't focus on the fun and focused solely on the tech, which always ends in disaster.

It was just a video of the whiteboard tech.

I don't think it's fair to say 3DR "focused solely on the tech."

[Jimmy]

TerminX, on 28 November 2014 - 03:55 AM, said:

Literally everything you just said is so far out of touch with reality that I'm seriously left wondering if your life has spiraled out of control with drug use or something. What happened to you?

Quoted just so I could say 15/10 great post, I laughed so hard.

MYHOUSE.MAP, on 01 December 2014 - 09:53 AM, said:

It sounds like to me that Broussard tried to play catch up with Carmack and it completely fucked the project over. It reminds me of what happened to Deus Ex: Invisible War, but on a much longer and larger scale (facny dynamic lighting system detroys the game) It's compeltely irresponsible managament to be making a dramatic change like that so far into development. If he hadn't tried to beat Carmack, then I'm guessing that DNF would've been released around 2004ish.

Someone on the 3DR Forums once said something pretty eloquent, and I'm paraphrasing; DNF was kind of like a guy trying to build a house while keeping up with a ship going down the river. Every time the ship got too far away, he decided to start building a completely new house instead of finishing the houses he started before.

[MYHOUSE.MAP]

High Treason, on 01 December 2014 - 10:16 AM, said:

I think the odds were somewhat against the announcement though to be fair, because Duke 3D was from a turning point in the industry. Go back a few months before that game and you could still make a game in less than a year/two years with a team of less than 20 people and be up to speed with the competition so announcing your next game in your current one was just the done thing. A few months later, you suddenly needed more time, more people and more money to get things done and this tradition disappeared.

Technology moved very quickly too; computers in 1996 were topping out with the Pentium 200 (Jun 96) and rarely had 3D accelerators, by 1998 this had more than doubled to 450MHz (Aug 98 - though rare) as had the power of graphics accelerators due to nVidia starting to make decent stuff, ATi pushing the envelope with the Rage series and 3DFX doing whatever they did; mostly waste electricity. By 1999 the world had reached around 750-800Mhz and would quickly accelerate past the 1GHz barrier in 2000. This is only taking into account clock frequencies in computers of the era, it is leaving out other innovations, APIs and whatnot. Imagine trying to keep up with that and with the reputation of your previous game to live up to... It'd drive you nuts.

Though I agree, there were very poor management decisions made and it wrecked the game.

This is true, but it was still irresponible to not only make mockups on the Q1 engine when they had no intention of using it, they also had the images appear as part of a front-page article on DNF. Showcasing prototypes like that is a very dumb thing to do.

The engine is also another issue. They blew a ton of money securing the Q2 engine, then swapped to UE1 several months later. If they had actually shopped around and found an engine that worked with what they wanted instead of just saying "must get juan carmake's engine bc he is very smart man", they wouldn't have blown about $500,000. Hell, Epic was letting people license UE1 before Unreal 1 was released, though they only got the source after Unreal 1 was released. It would've been a great moneysaver if they demoed it and saw that it could do things like the Hoover Dam that the Q2 engine would struggle with.

DNF was fucked in it's early years because of a pincer attack of rapidly improving tech and not adjusting to the changes in game development. The game would've had a good chance after the tech stopped improving at a breakneck speed, but Broussard was more interested inkeeping up with the Joneses instead of making his own niche, like with what 3DR did with 2.5D FPS games and DN3D. He tried to mimic every popular thing at the time, from fancy-schamncy dynamic light renderers to Halo-style gameplay, which damned the game.

Though honestly, I think what happened to DNF turned out to be for the best. It sucks that Duke is stuck in Gearbox hell while GB churns out more bland and grindy Borderlands games with ~wacky~ plots and characters, but it lead to Fred and co. acquiring 3DR and starting a new era for 3DR instead of it rotting to nothing, getting trapped in an endless cycle of making vaporware, or being a part of endless half-baked schemes, like the "new" Apogee. The new 3DR has been doing incredible things, like the cool anthology of nearly every 3DR title they've released and their desire to release Apogee/3DR game prototypes, and they seem to have good plans for 3DR going ahead. I don't expect 3DR to be a big player again, like it was after DN3D was released, but it'll be nice to have a publisher/developer taking good care of Apogee's legacy while making new and fun games.

[Richard Shead]

MYHOUSE.MAP, on 01 December 2014 - 08:22 PM, said:

Though honestly, I think what happened to DNF turned out to be for the best.

For the best?!?! I do understand that some good has come out of it - such as the acquisition of 3DR by Fred - but I consider that to merely be the silver lining in the big, dark cloud that is DNF. I don't understand how you can say that the far-less-than-satisfactory product we received in 2011 is a representation of an ideal scenario. Had DNF come out circa 2002 and been stellar, 3DR could have made a sh!tload of $$ and begun working on their next big thing - be it Duke 5 or another IP such as Bombshell- and then continued on to pump out extraordinary, groundbreaking games as a hyper-competent developer with a flawless reputation. We'd have the Duke product that we all craved and deserved, and 3DR and its fans alike would have a bright future ahead of them. Win/Win.

That's just my take on the situation...but I'm still cautiously optimistic about the new iteration of Apogee and looking forward to seeing what this reality holds for them.

[MYHOUSE.MAP]

Duke Rocks, on 01 December 2014 - 09:11 PM, said:

For the best?!?! I do understand that some good has come out of it - such as the acquisition of 3DR by Fred - but I consider that to merely be the silver lining in the big, dark cloud that is DNF. I don't understand how you can say that the far-less-than-satisfactory product we received in 2011 is a representation of an ideal scenario. Had DNF come out circa 2002 and been stellar, 3DR could have made a sh!tload of $$ and begun working on their next big thing - be it Duke 5 or another IP such as Bombshell- and then continued on to pump out extraordinary, groundbreaking games as a hyper-competent developer with a flawless reputation. We'd have the Duke product that we all craved and deserved, and 3DR and its fans alike would have a bright future ahead of them. Win/Win.

That's just my take on the situation...but I'm still cautiously optimistic about the new iteration of Apogee and looking forward to seeing what this reality holds for them.

3DR was going to be screwed sometime down the line because of horriffic management. The writing was on the wall as early as Shadow Warrior's development, when the game was drastically retooled from a bland, but competent ninja game (as seen in v0.90's unused maps) to Asian DN3D for little reason even though, according to online updates and the state the 0.90 maps were in, most of the old game was completed and was originally at 3DR to wrap it up instead of completely retooling it. Becaue of this, development was dragged past the point where 2.5D games were dead in the water and it sold poorly when it was finally released. George revealed Lo Wang's name to the public without informing any of the other developers about the name's existence before-hand, which is very, very stupid.

The interview that 3DR did with Mike Maynard also seems to hint at poor timing from 3DR, as their suggestions for the first Blake Stone were made much later than they should've. One of the reasons that iD didn't partner with Apogee for Doom was Apogee's poor ordering system, which was just some bored part-time college students screwing around. Both of these suggest to me that 3DR had serious issues even as far back as the tail end of Apogee's glory days that eventually manifested itself as major management issues that killed DNF.

In happy ideal land, 3DR would still be alive and making great games, but looking at 3DR's past and what happened to DNF, they were fucked because of their poor management. It was going to bite them in the ass sometime down the line, and when game development became much larger and expensive than back in the DOS days, recovering from bad management became very difficult, especially when your main manager still seemed to think he was developing DOS games back in the early 90s instead of a blockbuster PC game in 2002.

[OpenMaw]

Duke Rocks, on 01 December 2014 - 09:11 PM, said:

For the best?!?! I do understand that some good has come out of it - such as the acquisition of 3DR by Fred - but I consider that to merely be the silver lining in the big, dark cloud that is DNF. I don't understand how you can say that the far-less-than-satisfactory product we received in 2011 is a representation of an ideal scenario. Had DNF come out circa 2002 and been stellar, 3DR could have made a sh!tload of $ and begun working on their next big thing - be it Duke 5 or another IP such as Bombshell- and then continued on to pump out extraordinary, groundbreaking games as a hyper-competent developer with a flawless reputation. We'd have the Duke product that we all craved and deserved, and 3DR and its fans alike would have a bright future ahead of them. Win/Win.

That's just my take on the situation...but I'm still cautiously optimistic about the new iteration of Apogee and looking forward to seeing what this reality holds for them.

Agreed.

[Micky C]

MYHOUSE.MAP, on 01 December 2014 - 09:40 PM, said:

3DR was going to be screwed sometime down the line because of horriffic management. The writing was on the wall as early as Shadow Warrior's development, when the game was drastically retooled from a bland, but competent ninja game (as seen in v0.90's unused maps) to Asian DN3D for little reason even though, according to online updates and the state the 0.90 maps were in, most of the old game was completed and was originally at 3DR to wrap it up instead of completely retooling it. Becaue of this, development was dragged past the point where 2.5D games were dead in the water and it sold poorly when it was finally released. George revealed Lo Wang's name to the public without informing any of the other developers about the name's existence before-hand, which is very, very stupid.

The interview that 3DR did with Mike Maynard also seems to hint at poor timing from 3DR, as their suggestions for the first Blake Stone were made much later than they should've. One of the reasons that iD didn't partner with Apogee for Doom was Apogee's poor ordering system, which was just some bored part-time college students screwing around. Both of these suggest to me that 3DR had serious issues even as far back as the tail end of Apogee's glory days that eventually manifested itself as major management issues that killed DNF.

In happy ideal land, 3DR would still be alive and making great games, but looking at 3DR's past and what happened to DNF, they were fucked because of their poor management. It was going to bite them in the ass sometime down the line, and when game development became much larger and expensive than back in the DOS days, recovering from bad management became very difficult, especially when your main manager still seemed to think he was developing DOS games back in the early 90s instead of a blockbuster PC game in 2002.

That kind of reinforces the idea that Duke 3D was sort of a fluke. 3DR were certainly surprised by the extent of its success, and we all know that it wasn't the game they were working towards all development, rather was the last in a long line of continuous iteration.

[TerminX]

Yeah, Duke's personality and one-liners were literally added the month before release. The game had huge variation during development.

[MYHOUSE.MAP]

Micky C, on 01 December 2014 - 09:56 PM, said:

That kind of reinforces the idea that Duke 3D was sort of a fluke. 3DR were certainly surprised by the extent of its success, and we all know that it wasn't the game they were working towards all development, rather was the last in a long line of continuous iteration.

All games evolve as they're developed, but DN3D's development seems more like the devs were running around and screaming while trying out as many ideas as possible to see what would stick instead of having a concrete idea of what they wanted right off the bat.

[Micky C]

And then when they finally did against all odds find a winning combination out of the mess of different ideas, they decided to toss it all out for the sequel

[MYHOUSE.MAP]

Micky C, on 01 December 2014 - 11:04 PM, said:

And then when they finally did against all odds find a winning combination out of the mess of different ideas, they decided to toss it all out for the sequel

This has been said before, but I don't think 3DR themselves understood what made DN3D a hit when they were developing DNF, at least during the 2000s. They kinda understood that interactivity was important, but they implemented it in a completely different way from how DN3D has it that made it into a gimmick instead of doing things like changing how maps play.

[Jimmy]

DNF had the same type of development, and 3DR learned it wasn't the 90's anymore.

[Jblade]

I don't agree that Duke3D was a fluke, I think 3DR were fairly talented (because SW was great even if it did just develop the Duke formula but they knew what stuff to add) and Broussard knew how to make a fun game but I think once they moved out of the garage developer mindset into the proper AA studio they struggled to match the workflow.

[Frederik Schreiber]

Great, insightful thoughts in this thread.

[MYHOUSE.MAP]

Jblade, on 02 December 2014 - 12:03 AM, said:

I don't agree that Duke3D was a fluke, I think 3DR were fairly talented (because SW was great even if it did just develop the Duke formula but they knew what stuff to add) and Broussard knew how to make a fun game but I think once they moved out of the garage developer mindset into the proper AA studio they struggled to match the workflow.

3DR was certainly talented, but the problem was that they weren't very organized. Instead of having a plan, it's like they were flailing their arms around and running around like idiots. Something like the Devastator should not have been worked a month before the game was released. It should've been done a long time ago or never got off the ground because it was thought of too close to release. Several of E2's levels were done last minute as well, which is mentioned in the guidebook; another sign of poor planning.

The large amount of phases that DN3D went through, like going from simply DN1 or 2 in 3D (Fred's video), expanded DN sidescroller in 3D (Lameduke), FPS with weapons copied straight from Doom and lots of interactivity, to the final game, tells me not all was well in the house of Duke when DN3D was being developed. That's not even including the legion of content that was implanted then removed, like the flashlight, the beer, the Flamethrower, the old Overlord, the skybox seemingly being swapped every few months, and Stadium reverting to a map made in 1994 at the last minute.

This wasn't limited to 3DR either, as seen with Quake 1's development, but the key difference is that other developers grew out of it, while 3DR got worse about it.

[oasiz]

Duke3D's somewhat chaotic approach again just reflects much on what HL2's development was.
The leak that happened had over ~1300 .vmf files (map sources), about every single area that ended in the retail game existed in some other form elsewhere or was heavily adapted from some existing concept.
It's ridiculous to think how much it evolved and changed all around before finally settling with that we have.

I actually took the time to go trough all of the maps and I can safely say that there is about enough content for two distinct games that share a similar setting but have a very different end result.
This is much like comparing Retail duke vs. Lameduke, except with HL2's case they had more than twice the assets that ended up in the final game (plus later state of development).
Granted, a lot of it was stuff like architechture galleries where you had multiple blocks of c17 street design as reference (that never appeared in the final game as-is) or various A.I. tests.
Still, about every corner in that game has a story, this is also why I find that game very fascinating to explore in game development perspective.
If Duke3D is anything like what I think it is, we are going to see a lot of weird stuff and gimmicks that got tried for a while before getting abandoned, might even have reoccurring level bits that they tried to fit in with a lot of scenarios (while re-skinning as needed to fit the theme), really fun to see where some things originated from.

When putting the two games in to scale, I'd say that HL2 suffered about as much from trial & error.
Can't wait to see what Duke has in store.

[Jblade]

Good points, but I think if that Duke3D and SW were the products of sloppy development they still turned out well enough but I think it was only sustainable with earlier levels of graphic fidelity. As soon as DNF started the speed and cost of development started to ramp up pretty quickly and their old methods just weren't cutting it when actual 3D came in full force.

[Shaq Fu]

Frederik Schreiber, on 02 December 2014 - 12:51 PM, said:

Great, insightful thoughts in this thread.

It's amazing man. I keep thinking over how one of my all-time favorite games was an accidental masterpiece.

[MYHOUSE.MAP]

I think the difference between DN3D and HL2's development is that it seems that the HL2 team had a fairily good ideas about things like enemy design right off the bat, but the DN3D team radically shifted ideas at various points throughout development. It seems like they hadn't even settled on the guns until mid 1995, if TerminX is correct about the weapons originally being copied from Doom. In Duke, even the simple "grunt" enemy overwent a major overhaul from "dude in spacesuit with pistol" to "Dude in spacesuit with a giant crest and a pistol" to "lizard man that looks like a cat at times that has a laser pistol and has various recolors". It seems to me the only consistent things about DN3D through the various phases are the environments, a handful of the weapons, and one of the enemy types.

Jblade, on 02 December 2014 - 01:45 PM, said:

Good points, but I think if that Duke3D and SW were the products of sloppy development they still turned out well enough but I think it was only sustainable with earlier levels of graphic fidelity. As soon as DNF started the speed and cost of development started to ramp up pretty quickly and their old methods just weren't cutting it when actual 3D came in full force.

True. Back in the day doing something like adding a weapon at the last second would work becuase of how simple it was and things like actually producing physical copies wasn't hard. However, the problem is that these days, game developing is a lot harder and you can't do things like add shit at the last minute because of how complex the game is now, the need to publish a ton of copies months in advance, and marketing needing a near-final copy for things like tours. 3DR simply couldn't adapt to developing like that, which was evident with Prey, and, to a lesser extent, Shadow Warrior. DNF was just the final nail on the coffin.