Diedel - what do you do again?

All non Descent related stuff. Keep it civilized, this is not an NHB area.
Sirius
Posts: 1990
Joined: Tue Sep 25, 2007 10:29 pm
Location: Bellevue, WA
Contact:

Diedel - what do you do again?

Postby Sirius » Sun Jul 26, 2009 1:03 am

Not intending to get any information off you that you don't want to give. :) But I've seen you refer to some kind of mysterious (to me at least!) programming work, and I'm somewhat curious; what kind of work is it? If you're like most people you probably don't program games for a living! But I understand you have been working in the field for many years.
User avatar
karx11erx
D2X-XL Master
D2X-XL Master
Posts: 8112
Joined: Mon Sep 24, 2007 8:48 pm
Location: Wilferdingen, Germany
Contact:

Postby karx11erx » Sun Jul 26, 2009 9:08 am

Until two or three years ago I was researching, implementing and maintaining methods for fast pathfinding in (road) networks (applied in GPS based navigation system among others). I also did some customer related products around this.

Part of my job is to find or develop and evaluate new technology for future use in the company I work for.

Latest project is application of Stream Computing/CUDA and OpenGL rendering to speed up processes running on our servers and even on small devices (which more and more are equipped with hardware rendering support).

Another part is of my work is to optimize existing software by streamlining aspects of their implementation, using better tools like the highly optimizing Intel compiler, threading them into the companie's build processes, or automating processes like data processing deploying custom built scanner/parsers (using antlr and DataScript), allowing to automatically verify data formats, data integrity and provide data access by simply deriving it from an abstract data definition.

Yet another part (that is somewhat slumbering currently) is quality management (which is the level above quality assurance). The task of quality management is to determine the needs of a dev team, department or company in regard of quality assurance (processes, tools, ...) and provide these so that the quality assurance can work efficienly and derive meaningful quality measurement data from their processes and products (like number and severity of problems and trends in occurrence of such problems ...)

Before I started at my current employer I had a small company of my own for about two years, where I was selling a text based window and menu management system, quite successfully, I have to say. The advent of Windows 3.11 rendered it obsolete though.

Besides that have I always been doing private projects in my free time. I started with a mixed assembly language/PASCAL game of life implementation way back on a Apple ][e in the first semester of my study at the University of Karlsruhe (the editor was PASCAL, evolution engine was 6502 assembly code), wrote a database program to support playing those old-style text based dungeon games on a DEC VAX/VMS machine, implemented a computer supported PBM (play by mail) multiplayer game (similar game was conducted by a commercial company who sued me for it ... :P - didn't make money of it though, only asked for people to pay P&P). Later I developed a battle calculator for the MMOG Planetarion and finally ended up with DLE-XP and D2X-XL which really are quite enough not to leave room for much new stuff. I am planning to look into Unreal 3 engine modding now though with the goal of creating a truly modern Descent inspired game (actually a clone as much as possible and reasonable: Up-to-date graphics, Descent 2 gameplay, weapons, general physics).

My main development platform is MS Windows, but I have worked on main frames, midis, UNIX machines and even IBM's AS/400 systems too, running IBM/VMS, VAX/VMX, HP-UX, Sun Solaris, OS/400 ... which finally made me want to work with an OS that felt somewhat more like a "real" OS than MS Windows, so I started to fiddle around with Linux (in my leisure time again).

All in all I am an IT pro for over 20 years and have total coding experience for 30 years now (I started on a Commodore PET 2001 at school when I was 17).
Last edited by karx11erx on Fri Jul 31, 2009 9:12 am, edited 1 time in total.
Sirius
Posts: 1990
Joined: Tue Sep 25, 2007 10:29 pm
Location: Bellevue, WA
Contact:

Postby Sirius » Sun Jul 26, 2009 9:32 am

That sounds really cool. :) The "hard" stuff always fascinates me the most, since it's pretty easy for most people to do simple algorithms, or tools, or whatever - but things like optimisation, and programming close to the hardware, isn't something everyone can do.

I have a little bit of a "dream" of my own of making levels for UT3, which I think I told you about ... once I finish my current level for DFW I think I will start taking it up more seriously since the D2 engine is running out of steam a bit. You've done great work extending it, but I do sense that things like scripting, more exotic geometry, and hi-res robot animation may be beyond your desire to invest time in the game! :)
User avatar
karx11erx
D2X-XL Master
D2X-XL Master
Posts: 8112
Joined: Mon Sep 24, 2007 8:48 pm
Location: Wilferdingen, Germany
Contact:

Postby karx11erx » Sun Jul 26, 2009 9:35 am

Indeed. I'd rather invest that time into exploring a real good game engine offering all that and implementing well polished Descent 2 game play on top of that. D2X-XL is pretty much done. I really don't see much sense investing significantly more time in it.

I am not a fat, bald headed computer geek though ... :roll: I spoken about some of my sports activities here in the past already. Learned snow boarding a few years ago. Maybe I'll start Taekwondo together with my son when he turns six. I'd really love to (did Judo for a few years when I was a teenager ...)
Woolie Wool
Posts: 19
Joined: Sat Jul 25, 2009 5:23 am

Postby Woolie Wool » Mon Jul 27, 2009 1:44 am

karx11erx wrote:Indeed. I'd rather invest that time into exploring a real good game engine offering all that and implementing well polished Descent 2 game play on top of that.


The problem with writing your own 6DOF game is that you can code the engine but you still need the actual game, which makes building the engine seem trivial by comparison. With modern technology you need hundreds of people and millions of dollars to create a polished product. If you're going to try to do it all yourself the original D2 is probably the best you can hope for. And in these times, with a failing economy and a PC gaming market largely uninterested in anything besides "casual games" (glorified newspaper puzzles), Generic Tactical Shooter 7: Special Operations, and The Sims: Gravy Train, an endeavor to launch a startup company to build a Descent-like game would be extremely unlikely to succeed. Hate to be discouraging, but the difficulty of making the actual game assets would make developing a game with advanced graphics and other technologies virtually impossible unless you're already a fairly sizable corporation or have major backing from a publisher, and if you have the latter the publisher can essentially tell you what game to make and you'll have no creative freedom at all. :(

Imagine trying to create a modern Hollywood blockbuster action movie. Making a high-quality game is just as huge an undertaking, and just as ruinously expensive. We're talking millions, even tens of millions of dollars/Euros, and you'd need hundreds of thousands of sales just to break even. Even if you raised the money, could you find 500,000 people willing to buy a Descent clone so you can get that money back before your investors and creditors make you their slave for life?
User avatar
Aus-RED-5
Forum Moderator
Posts: 1490
Joined: Tue Sep 25, 2007 9:02 am
Location: Adelaide, South Australia
Contact:

Postby Aus-RED-5 » Mon Jul 27, 2009 3:02 am

You don't know Diedel very well. hehe :P

If I'm not mistaken - from past conversations, he's not talking about buying a game engine, but rather buying a game with a good engine to MOD a Descent like game for all to use.

Not something to sell. ;)
----System Specs----
Case: Corsair Obsidian 650D Black
Motherboard: ASUS Sabertooth X58
CPU: Intel Core i7 960 3.20ghz
Heat Sink: Corsair H100 CPU Water Cooler
Memory: Corsair 6GB (3 x 2048mb) DDR3 1600 Dominator (PC12800 - TR3X6G1600C8D)
Video Card: ATi SAPPHIRE Radeon HD 7970 3GB GDDR5 OC Edition
Hard Drives: SSD Corsair Force Series 3 F120 120GB & WD5003ABYX Enterprise RE4 500GB
Media Player: LG BH10LS30 BluRay Burner
Power Supply: Corsair HX-850
OS: Windows 7 Pro 64Bit
Monitor: Benq XL2420T LED
User avatar
Obi-Wan Kenobi
Posts: 262
Joined: Mon Jan 26, 2009 9:56 pm
Location: Tatooine
Contact:

Postby Obi-Wan Kenobi » Mon Jul 27, 2009 3:06 am

hey Diedel maybe a special Physx map would be cool since Unreal Engine 3.0 does support this! heh imagine huge chunks of wall being blown appart and letting the slave GPU of your geForce SLI setup take care of the Physx needed for them effects, just an idea, though it would be cool to make it like Red Faction that you can blow up , crates, certain walls as such that lead to a secret area or so :)
Firefox: 1CPU 8 Core OpteronMP 6140 D1, 32GB PC3-12800 ECC Reg HP/Micron Server Ram, Supermicro H8SGL-B/F, Sapphire Nitro+ RX 580 SE 8GB GDDR5
Blue-Leader: 2CPU 6 Core OpteronDP 2435 D0, 32GB PC2-6400 ECC Reg Kingston Value Ram, Supermicro H8DAE-2, Sapphire Nitro OC+ R9 Fury 4GB HBM
Valkyrie: 2CPU 12 Core OpteronMP 6180 SE D1, 48GB PC3-12800 ECC Reg HP/Hynix & Samsung, Supermicro H8DG6-F, Sapphire Nitro RX Vega 64 LE 2.0 8GB HBM2
User avatar
Aus-RED-5
Forum Moderator
Posts: 1490
Joined: Tue Sep 25, 2007 9:02 am
Location: Adelaide, South Australia
Contact:

Postby Aus-RED-5 » Mon Jul 27, 2009 3:09 am

Obi-Wan Kenobi wrote:....though it would be cool to make it like Red Faction that you can blow up , crates, certain walls as such that lead to a secret area or so :)

There is something like that coming out soon.
Mine Wars
----System Specs----
Case: Corsair Obsidian 650D Black
Motherboard: ASUS Sabertooth X58
CPU: Intel Core i7 960 3.20ghz
Heat Sink: Corsair H100 CPU Water Cooler
Memory: Corsair 6GB (3 x 2048mb) DDR3 1600 Dominator (PC12800 - TR3X6G1600C8D)
Video Card: ATi SAPPHIRE Radeon HD 7970 3GB GDDR5 OC Edition
Hard Drives: SSD Corsair Force Series 3 F120 120GB & WD5003ABYX Enterprise RE4 500GB
Media Player: LG BH10LS30 BluRay Burner
Power Supply: Corsair HX-850
OS: Windows 7 Pro 64Bit
Monitor: Benq XL2420T LED
User avatar
Obi-Wan Kenobi
Posts: 262
Joined: Mon Jan 26, 2009 9:56 pm
Location: Tatooine
Contact:

Postby Obi-Wan Kenobi » Mon Jul 27, 2009 3:27 am

Aus-RED-5 wrote:
Obi-Wan Kenobi wrote:....though it would be cool to make it like Red Faction that you can blow up , crates, certain walls as such that lead to a secret area or so :)

There is something like that coming out soon.
Mine Wars

Damn that's acool concept it does have a Descent type of feel as ihave seen the vids, cool game, it's on my hitlist now hehe
Firefox: 1CPU 8 Core OpteronMP 6140 D1, 32GB PC3-12800 ECC Reg HP/Micron Server Ram, Supermicro H8SGL-B/F, Sapphire Nitro+ RX 580 SE 8GB GDDR5
Blue-Leader: 2CPU 6 Core OpteronDP 2435 D0, 32GB PC2-6400 ECC Reg Kingston Value Ram, Supermicro H8DAE-2, Sapphire Nitro OC+ R9 Fury 4GB HBM
Valkyrie: 2CPU 12 Core OpteronMP 6180 SE D1, 48GB PC3-12800 ECC Reg HP/Hynix & Samsung, Supermicro H8DG6-F, Sapphire Nitro RX Vega 64 LE 2.0 8GB HBM2
Woolie Wool
Posts: 19
Joined: Sat Jul 25, 2009 5:23 am

Postby Woolie Wool » Mon Jul 27, 2009 4:11 am

Aus-RED-5 wrote:You don't know Diedel very well. hehe :P

If I'm not mistaken - from past conversations, he's not talking about buying a game engine, but rather buying a game with a good engine to MOD a Descent like game for all to use.

Not something to sell. ;)


Well the thing is that to make a modern game you have to sell it to recoup the millions of dollars you borrowed to make it.
Sirius
Posts: 1990
Joined: Tue Sep 25, 2007 10:29 pm
Location: Bellevue, WA
Contact:

Postby Sirius » Mon Jul 27, 2009 5:30 am

Woolie Wool wrote:If you're going to try to do it all yourself the original D2 is probably the best you can hope for.


Yeah, uh... no. You're quite right that we couldn't hope to complete a super-slick game that would actually look as good as UT3, but it's quite easy to do better than Descent 2 in that engine without even spending any more effort on it. You simply take the high-res textures we already have, do some quick BSP structures (which will let you create stuff a lot faster than D2 does), and you already have something that will probably top even the best D2X-XL levels (so far). Funny thing is it's also likely to run faster, since breaking to a new engine means you don't have to do the light-spamming that brings XL to its knees. That and Epic already spent millions optimising it for us. :)

So yeah. It's worth doing even if we can't make a perfectly pro job of it. It'll be better than what we've got at the very least.
User avatar
karx11erx
D2X-XL Master
D2X-XL Master
Posts: 8112
Joined: Mon Sep 24, 2007 8:48 pm
Location: Wilferdingen, Germany
Contact:

Postby karx11erx » Mon Jul 27, 2009 7:30 am

Yes indeed, I am gonna write my new game engine. I started yesterday night, and it's already half done. It will blow the Unreal 3 engine and Carmacks tech 5 engine right out of the water, or what do you think about full dynamic ray traced voxel rendering? If you have a webcam it will even create procedural hires textures from scratch on the fly by scanning your wall paper and seeing which environment you prefer to live in!

Of course, given the small user base of Descent 2, to make up for the cost one copy of my Descent 6 (I chose 6 for 6dof) will cost 9999 Euros (I was tempted to charge 10000 per copy, but want to avoid the psychological barrier). Hey I know, you hardcore Desent fans will happily shell out that peanut of a price for the greatest game that there ever was, is, and will be.

:mrgreen:
User avatar
DescentMax
Posts: 61
Joined: Tue Sep 25, 2007 8:50 pm
Location: Oklahoma, United States
Contact:

Postby DescentMax » Thu Jul 30, 2009 7:57 pm

im sry but jw what is this new game engine do? Is it a new game?
im not very good at this stuff. lol
User avatar
karx11erx
D2X-XL Master
D2X-XL Master
Posts: 8112
Joined: Mon Sep 24, 2007 8:48 pm
Location: Wilferdingen, Germany
Contact:

Postby karx11erx » Thu Jul 30, 2009 8:08 pm

Do my entire job (the AI will be awesome!), cook coffee, organize all my vacations and make a lot of money for me by trading shares and stuff.

And produce a holographic 3D Descent environment with virtual optical controls for all players. Including flight feedback signals which are directly feed into your brain via your optical nerve.

And it will scan your bank account data and make sure you pay dearly for playing it. :P
VladR
Posts: 62
Joined: Thu Feb 28, 2008 8:21 pm

Postby VladR » Fri Jul 31, 2009 8:55 am

Woolie Wool wrote:The problem with writing your own 6DOF game is that you can code the engine but you still need the actual game, which makes building the engine seem trivial by comparison.
Trivial in financial terms. But only relatively. I`ll explain later.

Woolie Wool wrote:With modern technology you need hundreds of people and millions of dollars to create a polished product.

This would be no multiplatform game. Pure PC version. That alone halves your dev team. Do you really think we need 127 minutes of lipsynched in-game videos ? For a Descent ? Pure Descent gamplay revolves around fights in thight corridors, not hours of static-render-quality mo-capped / lipsynched characters.
I know that you didn`t even mention these in-game videos. So, why do I mention it ? Because that`s what consumes majority of resources.
Once you put that out of equation you find out you need only a handful of level-design artists. But I`ll get into resources later.


Woolie Wool wrote:If you're going to try to do it all yourself the original D2 is probably the best you can hope for.
And how is that a bad thing ? Catering for a niche means providing that niche what it wants. And that`s a D2-style gameplay. No cut-scenes, no pre-scripted bullshit. Just pure action, as we all want.


Woolie Wool wrote:And in these times, with a failing economy and a PC gaming market largely uninterested in anything besides "casual games" (glorified newspaper puzzles), Generic Tactical Shooter 7: Special Operations, and The Sims: Gravy Train, an endeavor to launch a startup company to build a Descent-like game would be extremely unlikely to succeed.
It can succeed if it`s generating profit. That`s all. But more on that, later.

Woolie Wool wrote:Hate to be discouraging, but the difficulty of making the actual game assets would make developing a game with advanced graphics and other technologies virtually impossible unless you're already a fairly sizable corporation or have major backing from a publisher,and if you have the latter the publisher can essentially tell you what game to make and you'll have no creative freedom at all. :(
That is correct.


Woolie Wool wrote:Imagine trying to create a modern Hollywood blockbuster action movie. Making a high-quality game is just as huge an undertaking, and just as ruinously expensive. We're talking millions, even tens of millions of dollars/Euros, and you'd need hundreds of thousands of sales just to break even.
As I`ve explained above, this is not Uncharted-style of game. You don`t need tens of millions for development. Adequately, marketing budget doesn`t have to start at $2M, if the market is concentrated over few forums, thus almost whole target market can be informed about the game with minimum costs (e.g. $10.000 ad campaign).

Woolie Wool wrote:Even if you raised the money, could you find 500,000 people willing to buy a Descent clone so you can get that money back before your investors and creditors make you their slave for life?
Of course you wouldn`t. But that`s exactly the whole point. You don`t need to spend $10.000.000 to create a next descent. There`s no market for that.

Here`s a rough sketch of the resources needed:
1 Project Leader (art quality/consistency, final level testing/quality acceptance,organizing the work)
4 level artists (each spending 2.5 months working/polishing one level), thus creating a 20-level campaign in a year
1 2D/3D artist (ships/level props, textures/HUD)
1 Multiplayer programmer
1 Engine programmer
1 GamePlay programmer

Assuming you already have an engine/tools(editor), you would need these resources in a first year of development
- engine programmer (playing a role of a project leader for first year, since the engine is done, so he doesn`t have a backlog of 100.000 lines of code to write)
- gameplay programmer
- 2d/3d artist (for initial props/textures)
- 1 level artist (for first few levels)

Those are enough so that the whole workflow matures/stabilizes and remaining level artists can come on board to start producing the levels at full speed/productivity (i.e. no waiting till some feature is implemented inside editor or some critical bug/workflow issue is fixed).

This is 4 people, i.e. 4x12=48 man-months for initial production of a playable demo with 1-2 levels and final process workflow.

Next year, you need those additional people:
- 3 level artists
- multiplayer programmer
- project leader
This is 5 additional people, i.e. 5x12=60 man-months.
We also need to have those previous people for another year, so that`s 48 man-months.

Let`s sum it up : 48+48+60 = 154 man-months.
Let`s say that one person earns 1.500 EUR (e.g. in Slovakia). With taxes/social insurances overhead that`s 2.000 EUR company cost per man-month.
154 * 2.000 = 308.000 EUR for Development costs

Let`s break down additional external costs:
- accounting/finances - 24*300 = 7.200 EUR
- offices (rent+electricity+water) - 24*1000 = 24.000 EUR
- Sound/Music SFX (external contractor) - 10.000 EUR
- Bug-testing - 5.000 EUR
-------------------------------
TOTAL additional costs : 46.200 EUR

Now, let`s add the development costs:
308.000 + 46.200 = 354.200 EUR

Of course, you need talented people, not people who are used to working in 100-large teams. Someone who has programmed the 3D engine and several games just by himself (hint : me :-) ). The same goes for artists, where at least one has to be able to create quality textures besides meshes.

If the engine is open-source, e.g. costs zero, that`s the FINAL COSTS - 354.200 EUR.

That is VERY FAR even from a million buck.

Now, such a game can cost 30-50 EUR, it`s a niche game, so fans will gladly pay full price. You`ll get 47.50 after your payment provider gets his cut.

Now, 354.200 / 47.50 = 7.456 copies



I believe, it`s unrealistic to sell 7.500 copies via shareware. Unless you get to Steam, which is more of a hard-core base. Then again, there you`d be glad to have the price of 25 EUR, but the user base might compensate for it.[/b]
Last edited by VladR on Fri Jul 31, 2009 6:18 pm, edited 1 time in total.

Return to “Off Topic”

Who is online

Users browsing this forum: No registered users and 4 guests