[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/vr/ - Retro Games


View post   

File: 15 KB, 233x185, download (15).jpg [View same] [iqdb] [saucenao] [google]
6407634 No.6407634 [Reply] [Original]

Let's settle this once and for all: who was the bigger genius and which was the greatest single engine designer and coder of all time, John Carmack or Ken Silverman?

>> No.6407640

>>6407634
Carmack
You're forgetting that he did Quake and Quake 2 in addition to Doom. Not to mention the advancements in 2D PC graphics with commander keen.

>> No.6407645

>>6407640
Ok, perhaps a more fair comparison would be to their achievements in the early to mid 90s pre-Quake

>> No.6407650

>>6407634
I dont know but the latter one's engine resulted in much better games than that made by the former
>yikes cringe zoom zoom muh doom wads
Fuck you

>> No.6407654

>>6407634
I don't know how Carmack's code looks but Silverman's code in the Build Engine is an absolute fucking mess

>> No.6407665

>>6407645
>pre quake
build engine was developed and released around the same time as quake.

>> No.6407669
File: 1.48 MB, 1920x1080, tim_sweeney_epic_16-9.jpg [View same] [iqdb] [saucenao] [google]
6407669

>>6407634
I am you, but stronger than you both combined.

>> No.6407676

The best technical achievements are rarely from big releases though.
What's your favourite obscure and not even necessarily good game that did something amazing tech wise for its time?

>> No.6407681

>>6407665
There was nearly a year difference, September 1995 for the Build Engine and June 1996 for Quake.

>> No.6407683

>>6407669
At least Carmack didn't sell his soul to the Chinese.

>> No.6407685

>>6407676
Give some examples

>> No.6407727

>>6407654
To be fair, Ken was a teenager when he wrote Build though, right? Carmack must have been well into his mid-late 20s by Doom and had a lot more experience.

I remember reading somewhere that Carmack reached out to Den years later trying to get him to work for id and he turned it down. Seems like Ken could have done a whole lot more than make random voxel engines, but to each his own I guess.

>> No.6407748

>>6407634
Ken for me, easily. Build might be a mess underneath the hood, but most of the games using it feel tight as fuck. Same reason the Dark Engine is my favourite early 3D one.

>> No.6407853

>>6407669
Yeah, sure... He was writing 2D puzzle games when Carmack and Ken were hard at work changing the gaming world forever.

>> No.6407882

>>6407650
tsk tsk
you post in bad faith anon, you never intended to have a discussion if your preemptive insults are to go by

>> No.6407941

>>6407654

>I have spent a lot of time reading great source code. Having exhausted the insanely good idSoftware pool, the next thing to read was one of the greatest game of all time: Duke Nukem 3D and the engine powering it named "Build".
>It turned out to be a difficult experience: The engine delivered great value and ranked high in terms of speed, stability and memory consumption but my enthousiasm met a source code controversial in terms of organization, best practices and comments/documentation.
http://fabiensanglard.net/duke3d/index.php


compare with
http://fabiensanglard.net/doomIphone/doomClassicRenderer.php
http://fabiensanglard.net/quakeSource/index.php
http://fabiensanglard.net/quake2/index.php
http://fabiensanglard.net/quake3/index.php
http://fabiensanglard.net/doom3/index.php

>> No.6407947

>>6407941
Build's source code wasn't public - only the object code was licensed out. Therefore, it was never meant to be beautiful.

>> No.6408143

>>6407634
Carmack is the better programmer. Silverman is the better coder.

>> No.6408157

>>6407941
The book is such a great read.
The Doom source code is so easy to parse. I am barely acquainted with C, but I can easily find the head and tail of it.
It took me an hour just to figure out a simple function in Build, though. Ignore the function names and try to look up their purpose in the dev documentation instead.

>> No.6408253
File: 324 KB, 1364x768, sshock.png [View same] [iqdb] [saucenao] [google]
6408253

>>6407634
The guys at Looking Glass Studios were far ahead of both of them.

>> No.6408268

>>6408253

the system shock engine is just a modified ultima underworld engine and it's poorly optimized as fuck

>> No.6408274

>>6408268
They just required stronger hardware so they could render proper 3D environments

>> No.6408282

>>6407634
Dont know about genius, but I fucking love Duke Nukem, its almost a perfect game. Then again it's a bit later than doom.

>> No.6408297 [DELETED] 

https://tcrf.net/Proto:Blood/Alpha_Demo#GEORGE.TXT
Required reading on Silverman and his engine.

>Now in June, he hacks in his "group file" system, and says we have to use it.
In one sense he's right -- he's hard coded the thing into his engine, and it's
broken Blood unless I can figure out which parts of the engine I need to
replace. By the time we're done with Blood, we probably won't have any of
Ken's engine left... Perhaps you'll want to reconsider the royalties that
you are paying him.
>Ken understands algorithms, and he understands graphics. What he doesn't get
is how to design SYSTEMS. Everything he does to the engine is a piecemeal
enhancement to something that should have been designed right from the
beginning. The memory system is a poignant example of this. It is a hack,
like most of his code, and not very robust. From what I've disassembled of
the group file code (trying to figure out a way to eliminate or replace it),
it too demonstrates plentiful opportunities for crashing the system --
pointers being freed without validation, files being closed without
verifying the handle, etc.
>I sometimes wonder, though, if I'm going against the flow. Perhaps Ken added
his cache system in response to other teams request for a memory manager?
Maybe the group file code was asked for by the Duke Nukem team? These fears
usually turn out to be unfounded, since other teams report similar problems
as a result of Ken's "updates."

>> No.6408306

https://tcrf.net/Proto:Blood/Alpha_Demo#GEORGE.TXT
Required reading on Silverman and his engine.

>Now in June, he hacks in his "group file" system, and says we have to use it. In one sense he's right -- he's hard coded the thing into his engine, and it's broken Blood unless I can figure out which parts of the engine I need to replace. By the time we're done with Blood, we probably won't have any of Ken's engine left... Perhaps you'll want to reconsider the royalties that you are paying him.
>Ken understands algorithms, and he understands graphics. What he doesn't get is how to design SYSTEMS. Everything he does to the engine is a piecemeal enhancement to something that should have been designed right from the beginning. The memory system is a poignant example of this. It is a hack, like most of his code, and not very robust. From what I've disassembled of the group file code (trying to figure out a way to eliminate or replace it), it too demonstrates plentiful opportunities for crashing the system -- pointers being freed without validation, files being closed without verifying the handle, etc.
>I sometimes wonder, though, if I'm going against the flow. Perhaps Ken added his cache system in response to other teams request for a memory manager? Maybe the group file code was asked for by the Duke Nukem team? These fears usually turn out to be unfounded, since other teams report similar problems as a result of Ken's "updates."

>> No.6408313

>>6407676
Panorama Cotton is unironically more impressive than Doom

>> No.6408740

>>6407669
sweeney is a cock sucker sellout

>> No.6408767

>>6407853
Cope detected. Guess who makes the industry leading game engine right now and didn't have to go to work for fucking facebook.

>> No.6408875

>>6407654
Build engine is a lot more optimized than Carmack's engines though. Doesn't matter if it's a mess.

>> No.6408905

>>6407640
Quake could very barely run on i486DX PC with a beefy video card. 5th gen consoles couldn't run them. Duke 3D could run on any machine that could play Doom.

>> No.6408954

>>6408143
>programmer
>coder
explain please.
i thought these 2 are the same.

>> No.6409081
File: 17 KB, 769x225, 2020-03-24 11_59_48-Window.png [View same] [iqdb] [saucenao] [google]
6409081

>> No.6409087
File: 24 KB, 300x400, 2-1.jpg [View same] [iqdb] [saucenao] [google]
6409087

This guy.

>> No.6409131

>>6409081
>tfw almost 30

Shit.

>> No.6409162

>>6409087
he couldn't even finish one game before becoming a normie, no thanks.

>> No.6409179 [DELETED] 

>>6408268
>>6408274
They're the Unity Engine of current year. It's cool but it's not that awesome. At least not after DOOM was released.

>>6407634
Voxels COULD HAVE BEEN revolutionary if GPU makers adopted. Well, they didn't . Too bad.

Ken was pretty halfassed about being a progger. He basically quit to go into academics.

>> No.6409204 [DELETED] 

>>6408313
Impressive for a consolefag railshooter.
That's like comparing Crash Bandicoot to any FPS that on PC.

>> No.6409206

>>6408954
Good god, the amount of people who would chew you out if this were /g/.

>> No.6409225

I'm sure many of you faggots will call me a LARPer, but fuck it, I will say it: I've met Ken once by chance.

Back in 2010, I was a grad student at Brown (CS). A good old friend from high school (let's call him Joe) was also a student at Brown, being advised by Prof. Harvey Silverman at the EE department.

Since the Engineering and CS buildings were next to each other, I'd stop at Silverman's lab from time to time to invite Joe for lunch. In one occasion, I took notice of this weird, pale guy in his mid-30s sitting in the back of the lab, furiously typing on his terminal. Since that was none of my business, I grabbed Joe and headed to the nearest Subway for some baloney.

As we ate our sandwiches, my friend briefly mentioned that the chap back in the lab was Prof. Silverman's son and he was there to help with some simulation for a research project. Joe also mentioned that Harvey's son was regarded as a wiz programmer in the lab, albeit a little aspergy. Finally, he briefly remarked that this guy had developed "a somewhat famous game engine for weird, violent games" back in the 90s. It took me a while to realize that he was talking about THE Ken Silverman.

By the time we were back in the lab, Ken was gone and I had missed my chance to talk to him. I sent him an e-mail later that day, but never got a reply.

>>6408767
You sound as if Carmack was forced to work at Facebook, when he actually went there looking for new challenges.

>> No.6409232

>>6409225
Sorry about the leddit spacing. I was writing some stuff in LaTeX and forgot to switch modes.

>> No.6409265

>>6409225
The professor's son, talk about jew nepotism
Brown is the shittyist of the ivys.

>> No.6409284

>>6409225
Hearing about people running into /vr/ figures is always interesting.
My elective film history professor designed Dark Seed 2 and co-designed Ultima VII with Lord British.

>> No.6409306

>>6409265
I wish I could post things like this without getting banned.

>> No.6409350

>>6408143
>Carmack is the better programmer
Lolno, he was nothing but a coder while romero american mcgee and others did the rest.

>> No.6409354

>>6409265
>nepotism
Are you saying that Ken isn't a genius and only stumbled there through connection?

>> No.6409364
File: 282 KB, 648x786, 1575766772154.png [View same] [iqdb] [saucenao] [google]
6409364

>>6408905
Quake was fully 3D with proper room over room. It's like saying Wolf3D is better because Doom cant run on a 286
Look I think Quake sucks ass in everything that isnt graphical prowess (for the time.) I will still say that overall it was a technological leap over anything Ken has done aside from maybe the new-ish Build 2 stuff.

>> No.6409368

>>6408954
programmer is someone who programs and a coder is a retard.
You program in code, you don't code code. It's like calling a chef a fooder or a baker a breader.

>> No.6409386

>>6409364
>It's like saying Wolf3D is better because Doom cant run on a 286
My point is Duke 3D ran almost as fine as Doom, and that's an incredible optimization.
>Quake was fully 3D with proper room over room.
Unlike building a doom level in wolf 3D engine, I don't think you couldn't build a quake level in Duke 3D. Build engine is more efficient in regard of level design. I don't know if voxel enemies would significantly slow down the game though but 3D enemies wasn't what Ken had in mind when he coded the engine.
>I will still say that overall it was a technological leap over anything Ken has done aside from maybe the new-ish Build 2 stuff.
Ken's Voxlap engine is more impressive than anything Carmack did.

>> No.6409398

>>6409386
>Ken's Voxlap engine is more impressive than anything Carmack did.
You're forgetting that Quake 3 has unironically been the basis for like most games after its release, right? That's probably wrong but it had a ton of influence for how games are made nowadays, including the popularization of OpenGL.

>> No.6409404

>>6409398
>You're forgetting that Quake 3 has unironically been the basis for like most games after its release, right?
Quake 3 didn't have fully destructible environment and objects.
>it had a ton of influence for how games are made nowadays, including the popularization of OpenGL.
Only because it was more popular.

>> No.6409412

>>6408767
>Guess who makes the industry leading game engine
Spoiler warning: not him personally. Besides, it's a totally braindead reasoning. Like saying the engine coder for Doom 2016 is a better programmer than Carmack because how much more advanced his engine is compared to the 93 Doom engine...

>> No.6409527
File: 705 KB, 640x3440, john_carmack_engines.png [View same] [iqdb] [saucenao] [google]
6409527

Carmack pulled off these first person perspectives that ran fast, there was pretty much no precendent for that
There are rumors that Carmack is some kind of alien, I haven't heard rumors of Ken being an alien

>> No.6409565

>>6409527
>no precendent for that
Carmack based his engine on a tech demonstration he saw at a software convention, it was used for Ultima Underworld. Also Midi Maze did it before both of them. Silverman is more impressive because he was only 17 when he made Ken's Labyrinth from ground up. Both Carmack's and Ken's engines were based on pre-existing game techs.
>There are rumors that Carmack is some kind of alien
His boss zucc surely is.

>> No.6409605

>>6409527
Don't waste your time, brainlets won't understand how the biggest part of the deal is not implementing a certain feature, but getting it to run at an acceptable framerate on ordinary hardware.

>> No.6409635

>>6409527
Huh. I 've heard he's an Artificial Intelligence wearing a human skin.

>> No.6409771

>>6409386
>Voxlap engine
Looked it up and dont understand how it works. I thought voxelgames cant have ceilings, like the old delta force games.
How do new voxel engines do it?
I mean a voxel - empty space - a voxel on top.
Any simple explanation?

>> No.6409783

>>6409771
Don't make it too comlicated. Voxel = VOlumetrix piXEL. Imagine them like lego blocks in a 3D space.

>> No.6409785

>>6409783
*complicated
*volumetric

>> No.6409793

>>6409783
So whats the downside to it, why not make all 3d games like that.

>> No.6409808

>>6409793
Voxels have to be really small to be able to make smooth geometry. But the smaller they get the more processing power they need. GPUs are optimized for polygons, therefore everything has to be done on CPU. Same goes for animation, instead of transforming vertices, imagine transforming hundreds of millions of individual voxels at the same time. Voxels are also not textured, but colorized (I'm not sure about the exact term), so building assets for a voxel engine is also quite complicated. Then there's also animation. And so on.

>> No.6409814

>>6409808
Forgot to add, Ken himself wrote a detailed explanation of problems related to voxels on his page, but I can't find it right now.

>> No.6409818

>>6409808
Hope they make a comeback. Destroying the ground in Z.A.R was awesome.

>> No.6409821

>>6409605
Quake 1 ran worse than dark forces 2, which had gargantuan levels.

>>6409793
More complicated than wireframe rendering and you cant put on pretty skins on those boxes, so AAA companies rarely use it.
Its an impressive tech though. This one is a more recent voxel engine https://youtu.be/aAgVSTrqNOc

>> No.6409915

>>6409821
>Quake 1 ran worse than dark forces 2, which had gargantuan levels.
DF2 didn't even have a software renderer.

>> No.6409976

>>6409915
Are you sure that you aren't confusing Dark Forces II with Jedi Knight II - Jedi Outcast?

>> No.6410008

>>6409976
Yes, I'm sure. DF(1) -> DOS, 2.5D; DF2 -> Windows, accelerated 3D

>> No.6410015

>>6409527
>I haven't heard rumors of Ken being an alien

The hint is in his surname.

>> No.6410080

>>6409398
Quake 3 was shit. It was made to compete against unreal but didn't have all the texture innovation so they just said lets speed it up

>> No.6410095

>>6410080
"Texture innovation" worth shit if it runs like shit - and that's exatly what Unreal Engine did. Every UE game during my childhood were nothing but a crash- and glitchfest with some story, while Q3 engine based games have always felt solid and smooth.

>> No.6410128

Duke3D was extremely, extremely buggy.
Try making a level in it and you'd realize.

>> No.6410143

>>6409976
Also, JK2 Jedi Outcast ran on the Quake 3 engine.

>> No.6410147

>>6410128
Working on one lately. What's the catch?

>> No.6410201

>>6410015
He's a Jew, not an alien

>> No.6410207
File: 10 KB, 225x225, 1588158059570.jpg [View same] [iqdb] [saucenao] [google]
6410207

>>6407941
>Upon opening VS, the engine source felt unwelcoming with difficult filenames (a.c, cache1d.c). Opening those files reveals something hostile to the eyes and the mind. An example among many others from Engine.c (line 693):

if ((globalorientation&0x10) > 0) globalx1 = -globalx1, globaly1 = -globaly1, globalxpanning = -globalxpanning;
if ((globalorientation&0x20) > 0) globalx2 = -globalx2, globaly2 = -globaly2, globalypanning = -globalypanning;
globalx1 <<= globalxshift; globaly1 <<= globalxshift;
globalx2 <<= globalyshift; globaly2 <<= globalyshift;
globalxpanning <<= globalxshift; globalypanning <<= globalyshift;
globalxpanning += (((long)sec->ceilingxpanning)<<24);
globalypanning += (((long)sec->ceilingypanning)<<24);
globaly1 = (-globalx1-globaly1)*halfxdimen;
globalx2 = (globalx2-globaly2)*halfxdimen;

>> No.6410219

>>6409081
Haha, based

>> No.6410224
File: 335 KB, 1280x762, reddit Spacing lmao.png [View same] [iqdb] [saucenao] [google]
6410224

>>6409232
don't validate people who complain about reddit spacing
it does not exist

>> No.6410225

>>6409225
Based as fuck story anon, but damn I wish you'd got to talk to him. Did you friend have any stories about him?

>> No.6410226

>>6410207
I'ts simply more compact than using braces and identation. Keep in mind this was developed under DOS, in 80x25 text mode. Besides, as I said above, only Ken used to have access to this code - and if he knew his way around it, it was good enough.

>> No.6410235

>>6410207
oh
ohhh
oh... :(

>> No.6410242

>>6410226
Why would you even do it like that to begin with even if it is for your eyes only is the question

>> No.6410246
File: 645 KB, 360x342, ooh.gif [View same] [iqdb] [saucenao] [google]
6410246

>>6410207

>> No.6410259

>>6409179
>They're the Unity Engine of current year.
Not true, Unity can't do anything that other high profile engines couldn't, however System Shock's engine could do much more than Doom's.

>> No.6410265

>>6410242
>why do you write your notes in your handwriting?
>why do you organize your files the way you are comfortable with?
>why do you prepare your food the way you like it?
>etc, etc.

>> No.6410270

>>6410259
Except for being true 3D... Later on, its core rendering system was used in Descent and Descent II.

>> No.6410302

>>6410224
Don't forget about all the retards who make every sentence a new paragraph that's an unreadable wall of text. No "leddit spacing" though so it's all nice and dandy 4chin certified.

>> No.6410346

>>6410224
>pwnage
Holy fuck thats cringe.

>> No.6410356
File: 1.73 MB, 1920x1080, Red Alert 2 Allied navy.png [View same] [iqdb] [saucenao] [google]
6410356

>>6408905
>Quake could very barely run on i486DX
No shit, 486's were 7 years old at this point. Cheap Pentium 75 and 90 were out for over 2 years at this point.

>5th gen consoles couldn't run them
None of them could run Duke3D properly either. Framerate chugged on all consoles.

>Duke 3D could run on any machine that could play Doom.
Bullshit. Doom ran great on a 486DX2, while Duke3D ran like ass on one.

>> No.6410419

>>6410356
>Cheap
No PC's weren't cheap back then, especially pentium, even when they're already out for 2 years they were still more expensive than consoles afaik.
>Framerate chugged on all consoles.
I remember total meltdown stuttering a bit on PS1, but not I would call chugged. Duke64 ran really good, it even had it's own custom full 3D explosion effects.
>Doom ran great on a 486DX2, while Duke3D ran like ass on one.
Here's both Duke and Doom on 486DX2. Both roughly look the same.
https://www.youtube.com/watch?v=BoD0KNEdXnY

>> No.6410458

>>6410419
>Here's both Duke and Doom on 486DX2. Both roughly look the same.

Check your eyes then. You can see D3D slow down even on the roof. Smooth D3D needs a Pentium or at least a 5x86.

>Duke64 ran really good, it even had it's own custom full 3D explosion effects.

It ran on a completely different engine, could not maintain a solid 30fps and skipped frames.

>No PC's weren't cheap back then, especially pentium
It was still the low end choice and the only 486 PCs sold at this point were AMD 5x86 spreadsheet boxes. No point in holding games back for 7 year old architecture.

>> No.6410471

>>6410458
So? Heretic doesn't run much better, if at all.

>> No.6410484

>>6410346
>complains about pwnage
>uses cringe as an adjective
Okay, zoomer.

>> No.6410667

>>6410458
>You can see D3D slow down even on the roof.
The roof is the most graphically demanding part of DN3D, and it wasn't a single digit slowdown. Doom had a slowdown too on the very first level there where there's literally nothing going on.
>No point in holding games back for 7 year old architecture.
Yes there is a point. It's called optimization. For a modern comparison nuDoom runs smoothly on a cheap ryzen 3. If it only ran on new i7 it wouldn't be praised for being a well optimized game.

>> No.6410678

>>6410201
What's the difference?

>> No.6410690

>>6410667
>If it only ran on new i7 it wouldn't be praised for being a well optimized game
Only if it didn't deliver.

All things considered, Crysis was a very optimized game for its time. Tons of games won't even run on dual-core configs today without any justification.

>> No.6410736

>>6410225
Not much. Ken mostly worked remotely and would only show up in the lab while visiting his parents in Rhode Island. He published a couple of papers and patents with his father on microphone arrays. He stopped collaborating once he joined Voxon Photonics to work on his voxel engine (2013 I guess?).

>> No.6410742
File: 206 KB, 400x400, unnamed.png [View same] [iqdb] [saucenao] [google]
6410742

>>6410207
>one liners
>no space between operator and operands
>no blank lines

>> No.6410753

>>6409225
>had missed my chance to talk to him
>sent him an e-mail later that day, but never got a reply
What if he ignore you when you talk to him?

>> No.6410845

>>6408306
This is the shit I come to /vr/ for, I honestly can't imagine being a game dev in the 90s with all of the horrible business practices and shit management. Also on that tcrf link
>Georges recommendation to defraud IRS

>> No.6410848

>>6410736
I just watched a presentation/lecture he gave on that 3D arcade machine thing he was working on, the VoxieBox.Actually looks really fucking cool, I don't know why I never heard of it.

Ken seems like a pretty big sperg though, but a smart one. He seems more sperg like than Carmack in front of a crowd honestly, even with that weird "-mmm" inflection Carmack does at the end of sentences

Link to the presentation:
https://www.youtube.com/watch?v=-q2zxXrtz1c

>> No.6410861

>>6410265
Ok, then why is he so retarded as to prefer such a mess?

>> No.6410873

>>6410861
Some people just aren't innately structured in their workflow, myself included.

>> No.6410964

Carmack, Silverman, Sweeney, etc. are the top elite of talented enthusiasts.

John had the most impact because he was practical, at the right position / team and had the people to change the gaming world forever. I'm forever grateful for Doom and Quake and their terse, yet elegant style.

Silverman seems to be the "most talented" as he doesn't need any aesthetics in his code... he just tackled a problem and then implemented it. His build code is a testament to how he does not need any verbal fluff in his writings. I really dig the inline asm routines in the build engine. A good base for learning the ins and outs of inline assembly when using Watcom C.

I never took a look at Sweeneys code but I remember reading interviews and he must be seriously clever to have pulled off the things he did.

A bit lower on the ladder come people like Abrash, Hecker & company who are very good at what they do but they lack the genius of formerly mentioned engineers.

>> No.6411017

>>6410964
>Abrash, Hecker & company
Who?

>> No.6411035

>>6411017
Michael Abrash
Chris Hecker

>> No.6411042

>>6407634
I like them both. I liked their work. They both are very relaxed persons. They both even answered my mails full of questions in the 90s.

>> No.6411050

>>6411042
Whoa awesome, what kind of stuff did you ask them? I wish we could read those letters anon

>> No.6411054

>>6407727
Carmack was 22 when he wrote Wolf3D and started work on Doom, completing it at 23.

>> No.6411063

>>6411050
Technical questions and about their education, background, future plans if I remember right. Maybe late 1996, when I was 15. I don't have any of the mails anymore.

>> No.6411070

>>6410346
It was a different time, zoomshit. It became cringe when your reddit friends used it in middle school.

>> No.6411108

>>6407882
Not him but come on now...

>> No.6411119

>>6407665
uhm no, earlier
https://www.youtube.com/watch?v=l8tDBg4pfYk

>> No.6411124

>>6411017

Michael Abrash was, is, a Very influential programmer in the field of optimization / rasterization on x86 hardware(albeit he worked also on other stuff).
He was one of Carmacks, and the whole PC scene's, most practical influences(e.g. ModeX, sprite / polygon rendering and later on highly specialized drawing functions in Quake + implementation / refinement of the PVS).
Read "Graphics Programming Black Book" and inhale some infamous facts of the "golden era" if you're up to it.

Chris Hecker wrote some "bleeding edge" articles on polygon rendering and physics back in the day and helped in various projects regarding rendering solutions.

A current enthusiast, who is damn good, is Fabian Sanglard - he wrote book on Wolfenstein, Doom and has very interesting articles on his website regarding various games and their technical realization. Highly recommended.

>> No.6411136
File: 940 KB, 976x904, 1569906238386.gif [View same] [iqdb] [saucenao] [google]
6411136

>>6411124
>Graphics Programming Black Book
ASM is too much for me.

>> No.6411143

>>6407634
The two made me realize that I'm an all-around average brainlet, so I stuck with art instead of coding.

>> No.6411208

>>6407727
Carmack is only about 5 years older than Ken. Not 10 like you're making him out to be.

Either way, they're both good and they both worked on engines that powered both of my favorite games, and both are also now working on pretty cool things.

>> No.6411230

>>6408306
>Demands:
>* 35% on sales of ruins
>* Severence
>* compensation for amounts that would have been made through ports

>Conclusion:

>We want to be fair, we want a clean exit

>We should to be free to get back on track with the project, but this may
not be possible due to the damage you have already caused. We feel that you
owe us at least this.

>Avoid any mention of personality conflicts, i.e. Greg Malone.

Goddamn, sounds like they really didn't end on good terms. Did they get their demands from Apogee/3D Realms?

Also the part about Ken makes him sound fucking insufferable. They paid to license that engine and he just tells them to fuck off when they need something fixed.

>> No.6411459

Carmack but he is a soulless code nerd who probably doesn't even like video games, for Silverman the code was just means to an end of what he wanted to play

>> No.6411538

>>6410207
I worked a little bit on Chocolate Duke 3D's code and can confirm it sucks, especially when porting to 64-bit. Lots of places where pointers are stored as integers, for example.

>> No.6411543

>>6411208
5 years is a really huge gap. Ken was only a highschool aged kid while carmack was a college graduate aged man.

>> No.6413204

>>6411538
Sounds like a nightmare

>> No.6413309

>>6411143

Wise choice.
Albeit I like C and Assembler, I'm more at home at making maps and playing guitar. One ought to have a certain mental sharpness, calmness and tenacity to be a successful programmer who can pull off the really nice stuff. Not everybody is cut out for it.

>> No.6413368

>>6413309
Same here. Came to realize just because I understand assembly, C, and a bit of C++, and I casually hack/play around with others' code, I will never be able to achieve anything under pressure, or even design something from the ground up within a given timeframe.

>> No.6413421

>>6408306
This might've ultimately led to Monolith developing their own engine in-house from that point on. 1 year later, Blood 2 was released not using Build anymore (outdated at that point too) but the first iteration of their own LithTech engine.

>> No.6413431

>>6407882
I already know that no matter what ill get those 4 responses listed. Trying to discuss Quake and Doom 2 always ends instantly like that

>> No.6413461

>>6413421
I think it was Shogo that used the very first version of LithTech, but doesn't really matter: Blood 2 (and the engine it uses) was a fucking mess, and it was a miracle that it even ran somehow.

>> No.6413890

>>6411230
Was Ken the cause of them being so pissed off and wanting to terminate their business relationship with 3D Realms?

>> No.6413901

>>6411459
>Carmack but he is a soulless code nerd who probably doesn't even like video games
This is a retarded meme, he was super into D&D, metal, scifi, horror flicks and Nintendo games. He also has a wife and son nowadays, he has said that he was "sort of an amoral little jerk" when he was younger because he knew he was smarter than everyone else. He has said he would have done some things differently in hindsight as far as iD as well, since his notorious coldness for anything other than programming (Masters of Doom recalls an anecdote where one of the guys at iD, can't remember who specifically, maybe Petersen, told Carmack his daughter had finally gotten into the prestigious college she wanted and which he had been talking about for months and just got a "Mmm" from Carmack) alienated alot of people back in the day

>> No.6413905

>>6413890
No, they most likely received a better offer from GT Interactive regarding Blood.

>> No.6413908

>>6413905
Well then what is with all the venting and plans to demand an out due to all the "damage" they had caused them, referring to wanting a "share of the profits of the ruins" here: >>6408306

Also the stuff they write about Ken really is unacceptable, can you imagine Carmack telling an engine licensee "I'm not fixing that bug, just go back to the old and broken way you were doing it"?

>> No.6413918

>>6413908
Read the entire documents (plust the other included development logs), not just that little excerpt regarding Ken. You'd see things weren't as bad, and it's mostly just a rage fuelled rant.

>> No.6413930

>>6413908
Also, Carmack gave out the full source code of everything (except the DMX sound engine, the only third-party code they used), so he didn't have to "fix anything for them". On the other hand, I don't know why Apogee/3DR kept the Build code as a trade secret up till 2000 or so, it really makes no sense.

>> No.6413937

>>6413908
Id software left Apogee because of shit business practices and awful contracts.
It seems like Apogee hadn't changed much and Monolith were seeing the same things.

If it really was just the engine, they would try to negotiate instead for more time and resources to build their own engine for the next game. Instead they wanted a full exit from Apogee.

>> No.6413941
File: 27 KB, 438x640, 447229-duke08.jpg [View same] [iqdb] [saucenao] [google]
6413941

>>6410845
>Nah, it's cool dudes, just commit tax fraud. You can even use my computer!

>> No.6413946

>>6413930
iD licensed Carmacks engines that weren't released open source yet dumb ass

>> No.6413949

>>6413461
Blood 2 was basically an unfinished game shoved out the door due to publisher pressure. They refined the engine for future titles like No One Lives Forever and it was much better.

>> No.6413950

>>6413930
You know damn well Carmack would and probably did provide support and documentation for his engines, whereas according to Monolith Build had no documentation and the support from Ken was "I don't know"

>> No.6413968
File: 156 KB, 140x553, bbag.png [View same] [iqdb] [saucenao] [google]
6413968

>>6413941
>Hey... You see this fucking bag? You like it? It's Shell Cordovan. It's full of fucking money, man. How do you think I got all this? By defrauding the IRS!

>> No.6413969

>>6413946
You are the only dumbass here. Check the original source code of Raven's games, you'll see everything is there (apart from DMX, as mentioned above). The Strife team also had access (and made modifications) to some core engine functions.

>> No.6413974

>>6413969
Talking about after that

>> No.6413979

>>6413949
>The original Blood had it's rights tied up for over a decade
>No One Lives Forever is currently in licensing hell
>Monolith has had a ton of bad breakups with publishers
Monolith seems like they had a hard time with legal law for their entire life.

>> No.6413980

>>6413979
So are Monolith the dickheads then?

>> No.6413982

>>6413950
You misunderstood the leaked Monolith rant again. The guy didn't say there was no documentation for Build. He complains that the implementation of a certain component is flawed.

>> No.6414323

>>6407634
According to Master's of Doom Silverman was influenced by Carmack a lot. They managed to get the two in contact with each other during the development of Duke 3D. It was because of this that Silverman made the Build engine sector based. Carmack would later view the Build engine as a "fucking mess" because much of its technology was essentially hacks stapled onto it.

Additionally the Build engine was the only thing Silverman really did. After that he tried making voxel based engines but didn't really do anything as big as the Build engine after that. By comparison Carmack after Doom made the Quake engine which was a huge technical feat at the time.

A better question is Carmack vs Sweeny since Sweeny was a savant as well but he was a better business type. (He focused a lot on building the Unreal engine so it was capable of being used by outside developers. To the point where they even wanted to just release the Unreal SDK by itself to consumers).
>>6407748
That wasn't due to Ken as much as the actual developers who worked on the games.

>> No.6414372

>>6414323
From what I gathered, Ken wasn't really interested in the gaming industry to begin with, and I could only assume being a part of it for a few years only made him even more disillusioned. So instead of coming up with something new after Build, he chose to be an academic. It's really not when people keep on parroting how "he did nothing beyond Build".

>> No.6414375

>>6414372
*really not fair

>> No.6414398

>>6407634
The one that didn't google 'how to destroy evidence'.

>> No.6414456 [DELETED] 

>>6410207
this look optimal actually. of course w/o documentation it would be harder to figure out unless your name is Ken sama!

>> No.6414463
File: 23 KB, 640x400, nene.asm.png [View same] [iqdb] [saucenao] [google]
6414463

>>6414323
>fucking mess
Did Ken make the Build engine alone?
Maybe he didn't have anyone whom he could discuss to improve his code.

>> No.6414497 [DELETED] 

>>6413982
I don't see why you are so intent on trying to cover for Ken. These problems were clearly more than just a single incident and if Ken's engine had proper support and documentation then such fundamental flaws as mentioned below wouldn't be unknown till then. Also the page anon linked to the former Amazon programmer who reviewed Builds code remarked there was a lack of proper and robust documentation as well.
>Everything with this system was going fine until we discovered that Ken has
some engine flaws in dealing with sprites that use origin based centering. °The clipmove() function doesn't work, and neither does hitscan(). These two
functions are pretty much essential to doing anything with the engine. I
could rewrite them, of course, but that would be a few days to weeks out of
our schedule, depending on how many more bugs and design problems crawl out of the woodwork.

>We had waited a few months before switching to the new origin based centering, so I was quite surprised to find such fundamental problems with it. We called up Ken and told him about it, and he said he didn't think anyone was using it, but he really didn't know. We asked him to fix it.

>Tonight I called up Ken because of the group file problems, and because we
still don't have a fix for the clipmove() or hitscan() problems. Ken he wasn't going to fix it and we should go back to using the old centering mode. This is totally and completely unacceptable. We've already invested time and effort (need I say money?) into using something, and we're not about to waste more regressing.

>> No.6414498

>>6407634

White anglo suburban ubermensch computer scientist vs Ashkenazi jewish degenerate "coder". Gee I wonder which is going to make the more net positive, impactful, and long lasting changes to what ever industry they apply their talents to?

>> No.6414504 [DELETED] 

>>6414497
I don't see why you are so intent on trying to cover for Ken. These problems were clearly more than just a single incident and if Ken's engine had proper support and documentation then such fundamental flaws as mentioned below wouldn't be unknown till then. Also the page anon linked to the former Amazon programmer who reviewed Builds code remarked there was a lack of proper and robust documentation as well.
>Everything with this system was going fine until we discovered that Ken has some engine flaws in dealing with sprites that use origin based centering. °The clipmove() function doesn't work, and neither does hitscan(). These two functions are pretty much essential to doing anything with the engine. I could rewrite them, of course, but that would be a few days to weeks out of our schedule, depending on how many more bugs and design problems crawl out of the woodwork.

>We had waited a few months before switching to the new origin based centering, so I was quite surprised to find such fundamental problems with it. We called up Ken and told him about it, and he said he didn't think anyone was using it, but he really didn't know. We asked him to fix it.

>Tonight I called up Ken because of the group file problems, and because we still don't have a fix for the clipmove() or hitscan() problems. Ken he wasn't going to fix it and we should go back to using the old centering mode. This is totally and completely unacceptable. We've already invested time and effort (need I say money?) into using something, and we're not about to waste more regressing.

>> No.6414509

>>6413982
I don't see why you are so intent on trying to cover for Ken. These problems were clearly more than just a single incident and if Ken's engine had proper support and documentation then such fundamental flaws as mentioned below wouldn't be unknown till then. Also the page anon linked to the former Amazon programmer who reviewed Builds code remarked there was a lack of proper and robust documentation as well.
>Everything with this system was going fine until we discovered that Ken has some engine flaws in dealing with sprites that use origin based centering. °The clipmove() function doesn't work, and neither does hitscan(). These two functions are pretty much essential to doing anything with the engine. I could rewrite them, of course, but that would be a few days to weeks out of our schedule, depending on how many more bugs and design problems crawl out of the woodwork.

>We had waited a few months before switching to the new origin based centering, so I was quite surprised to find such fundamental problems with it. We called up Ken and told him about it, and he said he didn't think anyone was using it, but he really didn't know. We asked him to fix it.

>Tonight I called up Ken because of the group file problems, and because we still don't have a fix for the clipmove() or hitscan() problems. Ken he wasn't going to fix it and we should go back to using the old centering mode. This is totally and completely unacceptable. We've already invested time and effort (need I say money?) into using something, and we're not about to waste more regressing.

>> No.6414515

>>6413918
Fuck off Ken

>> No.6414524

>>6414463
Carmack made his engines alone too, though he would've had other programmers like Romero to discuss it with, no one was on his level especially later

>> No.6414569

>>6414524
Romero and tom hall worked closely with carmack. Ken didnt have anyone like them and he had only graduated high school, he lacked experience working with a team.

>> No.6414583

>>6409915
>DF2 didn't even have a software renderer.
You moron. It's the 8bit mode, you can even run it on a fucking VGA.

>> No.6414595

>>6410458
>It ran on a completely different engine
Correction, it ran on a different renderer, whether or not it was completely different is up to debate. You're thinking Duke3D on Saturn.

>> No.6414607
File: 156 KB, 1024x761, reddit spacing time traveler.jpg [View same] [iqdb] [saucenao] [google]
6414607

>>6409232
>leddit spacing

>> No.6414647

>>6410207
I dunno it's fine. But I read machine decompiled c every day, so...

>> No.6414874

>>6407676
Most Traveler's Tales games for the Genesis

>> No.6415554

>>6414509
>I don't see why you are so intent on trying to cover for Ken.
You are the one who's furiously anti-Build for some reason, using the Monolith report as the sole source of your information. Given that, there's no use suggesting you to learn how Build actually is, or perhaps even get you to read its documentation, or code/reports done by anyone other than that single Monolith programmer.

>> No.6415558

>>6414583
>B-but it runs faster than Quake, with half the polycount!

>> No.6415607

>>6414607
>That font
>Those colors
>The check box has shadows
I wanna go back guys

>> No.6415727

>>6409527
could you turn the brightness up a touch on that last one?

>> No.6416060

>>6413368

Which makes Carmack's and Silverman's achievements all the more impressive.

>> No.6416073

>>6414323

Ken did some very clever things nonetheless - his engine also wasn't meant to leave a code legacy to "build" upon. If you haven't already read it - Fabian Sanglard wrote an awesome article about Ken's work on build.

From what I have read Ken said that he had lost the ball when the tech went full 3D and couldn't catch up. If one takes a look at Quake's source code it's understandable that he wasn't able to catch up in time... it's quite a complex undertaking solving pretty tough problems.

Do you happen to know if it's correct that Romero wrote the Quake C compiler? I remember Carmack being a bit pissed about the work Romero did on this one... but I couldn't find the .plan file.

>> No.6416079

>>6414569

Romero had most likely 0 input on essential things as he lacked the technical expertise.
He "only" did maintenance / side work on Doom(Ed)... which, granted, is most likely more than most people would ever be able to achieve.

>> No.6416086

>>6416073
I don't think Romero did any coding for Quake at all.

>> No.6416090

>>6415554

I think many of the devs that worked with Ken became hostile because they couldn't keep up with him - both positively and negatively speaking(talent and chaos wise).

They most likely had more experience but he solved, even if non-optimally, most problems by himself in his own style. I read a bit about the "Duke team's dev" diary and there's a bit too much bad "blood" in it.

When in doubt I go with the more talented person and that's easily Ken in this case.

>> No.6416102

>>6416086

Yes, it was too hard. But I'm not talking core technology but "helper" stuff.

I remember reading the Carmack, due to time constraints, shipped off some work to Romero and he returned him a wreck of a compiler. I remember reading this back in 98 or so on bluesnews...?

>> No.6416121

>>6407634
both are massive faggot thieves with the former being a known groomer

>> No.6416374
File: 122 KB, 500x397, anna-kang-daisy-may-trent-reznor-john-carmack-5-24-00-dallas-17439182.png [View same] [iqdb] [saucenao] [google]
6416374

>>6416121
Excuse me? What child has John Carmack been accused of grooming? The fucking guy didn't even have time for other humans until he met his equally nerdy digital artist/machinima making wife in 2000 who now runs a media company or something. Pic related, that's her.

>> No.6416382

>>6416079
I mean Romero was no slouch in the programming in the early days... He already had many published games under his belt when he met Carmack, who thought he was the coolest guy he had every met precisely for that reason. Now yes, to use a Dragonball analogy, Carmack went on to go Super Saiyan and Romero stayed as like... Yamcha. Or Maybe Krillin would be more fair.

But he did write a lot of the tools and the level editors and shit that were used on Wolf3D and Doom. I think his contributions became less and less by Quake.

>> No.6416398

>>6416102
Damn, I'd like to read that.

Googling it did give me this really cool article from WIRED in 1996: The Weird Egos at Id:
https://www.wired.com/1996/08/id/

>> No.6416403

>>6416398
-Weird, I guess I added that myself.

Weird.

>> No.6416413
File: 30 KB, 612x409, gettyimages-1156605148-612x612.jpg [View same] [iqdb] [saucenao] [google]
6416413

>>6416398
>But for most fans, there's nothing academic about id's games. Several Doom diehards who came to id one afternoon to perform beta testing demanded that Tim Willits customize their mouse and keypad configurations in accordance with highly specific instructions. Willits complied, commenting, "You know, none of us use these weird configurations."

>"Are you a Doom fan?" one beta-tester asked.

>Tim stared at him. "Uh, I work here, remember?"

>"No, man, you don't understand. Doom is my life!"

Hahahaha, that fucking sounds like what people from these threads would've said.

>> No.6416439

>>6416398
>https://www.wired.com/1996/08/id/

Dang, thanks a billion! I remember when this magazine came out skipping through it in the book shop at the train station. Didn't buy it though... finally I can read that thing in a peaceful setting.

Regarding Willits - a total turbofag, but he made some pretty solid maps and was employed at id when times where still cool.
Doom wise not all that hot but his map "The Edge" for Quake II was pure genius... guess he had a very good day when he came up with the concept.
He gets a lot of shit from the stupid community but he's at least half as good as Romero.

>> No.6416454
File: 8 KB, 360x270, 1349628-edg194_s_id_wilits.jpg [View same] [iqdb] [saucenao] [google]
6416454

>>6416439
Willit's has done some decent stuff, of course... but that whole mess a couple years ago with him claiming to the press he came up with the idea for official retail multiplayer only maps to be included with Quake and that Romero and Carmack both looked at him like he was crazy and said that was the stupidest thing they ever heard.

Enter Romero on his personal website and twitter, and later Carmack on twitter, to both say that incident literally never happened and Romero pointed out that no only had Tom Hall's Rise of the Triad and Marathon already come out with multiplayer only maps included, but Id themselves had released an official deathmatch map online, id Map01 for Doom II, by American McGee years before Quake.

Tim instead of walking it back and saying he misremembered instead posted a clip of some half-finished Quake map from 1996 he said was his WIP deathmatch map from that time and then doubled down and said he was sticking to his story.

I mean come on, Tim

>> No.6416461

>>6408274
>>6408268
>>6408253
Quake was the moment id had tech superiority over LGS.
They couldn't recover after this.

>> No.6416464

>>6416439
Oh you are very welcome for the link, I am really glad I could get it to you so many years after you saw it on the rack. It's weird how synchronicity works sometimes... I just happen to google a line in a post that catches my eye and you end up with an article from 20 something years ago you had wanted to reach in a train station.

>> No.6416502

>>6416454

Interesting - I listened to a recent interview of his and stated claims did not seem outlandish.
Well, if both Carmack and Bro-mero diss it then it's bullshit. He seems to have been a huge nerd who got lucky to land a job that granted him some fame... such people end up with a bit of delusional memory. I remember seeing the first Doom maps he sent in and they weren't all too hot... he was one of the people who were at the right time at the right place.

Nonetheless he worked hard and made up for his lack in genius when compared to Lawn-mero and McGee. The only thing that was really bad, was that he become creative director or something. I think a guy like him kills anything he touches because he lacks the genuine crazy that Romero has.

>> No.6416568

>>6416502
Romero's refutation is linked below. To be clear, Carmack only stated on Twitter "That didn't happen and I trust Romero's account of events." and didn't comment further. Tom Hall also chimed in about ROTTs multiplayer maps on Twitter.

Romero's post about Tim's claims and the facts:
https://rome.ro/news/2017/8/30/multiplayer-only-maps

>> No.6416584

>>6416502
>>6416568
A bit offtopic, but I wonder how Carmack and Romero gets along these days. I don't know how bad the situation was, all I remember is that the massive ego of the latter got on the nerves of the former around the final days of Quake's development.

>> No.6416594 [DELETED] 

>>6416461
>LGS.
thats not saying much since LGS was always running out of money.

>> No.6416642

>>6416584
From all the interviews I have seen, it's water under the bridge now... Romero and Carmack both speak well of each other and do keep nominally in touch. If either really needed a favor I am sure they could call the other, depending what it was.

Things got tense during the time but everyone seems over it now. Carmack for his part never seemed to get personally nasty about any of it.. it was just business for him. Basically "You aren't contributing, you need to go...". In those early days when they were all living in the lake house in Louisiana and working for Softdisk and staying up all night playing D&D, listening to metal and playing Nintendo, real friendships were forged, some like Hall and Romero endure to this day. But as things got bigger and bigger for them and the stakes got higher and higher due to their fabulous success, things changed and started to become business, especially for Carmack.

I will say I am unsure on the realtionshop with Tom Hall and Carmack. I have heard Hall speak fondly of the night he and Carmack stayed up all night making the Dangerous Dave in Copyright Infringement demo for Romero. But then when he starts talking about development of Quake, he starts sounding more and more bitter towards him. He claims the violence and dark aspects of the story had nothing to do with him leaving and that is revisionism - he says the real reason was one day Carmack just came in and said "We are doing one continuous world for Doom, no levels" and Hall desperately argued against it to no avail. So he came up with his story about saving your friend and started designing the world of his Doom bible, making it a long rescue story to accomdate Carmacks continuous game world. Then a few weeks later Hall came to work and Carmack said "Oh Tom, we are doing levels now." So he had all this work that was useless and had to be changed to fit the new format and he just started feeling discouraged and heartsick about the whole thing and eventually left.

>> No.6416647

>>6416642
*Development of Doom, not Quake, obv.

>> No.6416658
File: 32 KB, 610x344, 8rTR5YJ88zptkeb3ThbgEn-1200-80.jpg [View same] [iqdb] [saucenao] [google]
6416658

>>6416642
>n those early days when they were all living in the lake house in Louisiana and working for Softdisk and staying up all night playing D&D, listening to metal and playing Nintendo, real friendships were forged
Sounds so based, I remember reading about those days in Masters of Doom.

They are making a Netflix series based on Masters of Doom called "Masters". Anyone know the status of that? Here are their castings of Romero and Carmack. Gotta say the Romero guy looks a little weird but I assume they got him because he's Native American like John. The Carmack actor is just flattering.

>> No.6416752

>>6416121
Its unfair to judge people based on apperance but Carmack always looked like he fucked kids

>> No.6416760

>>6416658
>Masters of Doom
>Netflix adaptation
This is why we can't have nice things.

>> No.6416770

>>6416568
In January 2020, Willits was on the Arcade Attack Podcast and clarifies that when he talked about multiplayer-only maps he was specifically talking about Quake, not FPS games in general. He also added that Quake was the first FPS game that had dedicated client-server architecture for multiplayer.
https://en.wikipedia.org/wiki/Tim_Willits#Controversy

>> No.6416834

>>6416584

They are old farts now and have accomplished more than many could ever dream of - personally and financially. Silverman is, unfortunately, largely a forgotten obscurity while they became, rightfully, some small scale superstars.

Especially Carmack has always been a consummate professional, sleek but also very generous and on top of that laid back... too busy to grind an axe.

I wish that Romero had learned some humility from Carmack and kept more in the shadows.... more focused on productivity where it was needed - more high quality maps for DoomII.

>> No.6417040

>>6416770
He's still full of shit, as both Romero and Carmack said, he didn't come up with the idea for Quake.

>> No.6417045

>>6416760
Man, I just look forward to watching it with my wife since she could give a fuck less about Doom and has already had to hear more than she ever cared to about it from me over the years but the story of these guys is really remarkable and I think her and millions of other people who otherwise would never care will be captivated by it like I have been and that is good enough for me, regardless of if it's cringe or not.

>> No.6417059

>>6417045
*Maybe thousands is a more realistic hope, hundreds of thousands even.

>> No.6417123

>>6416584
It's pretty clear that Romero and Carmack aren't friends and don't talk outside of public events where the two of them meet. Romero and Carmack always had a mostly professional relationship in the 90s anyway. Carmack wasn't the type to play vidya with the bros after coding was done he would just seclude himself in his office and keep coding.

You can especially see this on twitter. The two post extremely regularly there and Romero will occasionally tag Carmack in a post, but Carmack never responds and never does the same. He generally only refers to Romero in passing.

Romero was besties with Tom Hall when working at Id and the two are still in constant contact today, they even started Ion Storm together.

>> No.6417209

>>6417123
>Carmack wasn't the type to play vidya with the bros after coding was done he would just seclude himself in his office and keep coding.
He has said himself he was a huge fan of Nintendo games and they all played everything in the office in the early days... He was the DM of the D&D campaigm that was the inspiration for the Demonic invasion of Doom and the character Quake was an ultra powerful NPC created and played by Carmack. To say he was all business is not accurate, he got that way later and wasn't as social as the others but he wasn't an unfeeling cyborg 24/7 (until maybe during Quakes development)

>> No.6418248

>>6417123

Romero seems like a kind of attention whore so no wonder Carmack largely has ignored his diva-esque persona. One has to only think of the cringe worthy sigil "package".
He has nonetheless earned some of his fame through talent and luck(his coding and Quake skills have always been vastly overrated though).

Hall's entries / stubs for Doom were very good.
He unfortunately seems to be even more tacky than Romero but can't pull it of because of his horrendous looks.

>> No.6418329

>>6407634
Carmack. If we just look at longevity and ease-of-use/modding, DOOM Engine/id Tech destroys Build.

>> No.6418527

>>6407669
https://tcrf.net/Jazz_Jackrabbit_(DOS)#Hidden_Messages

>> No.6418676

Do you think hey browse this board?

>> No.6418706

>>6418329
Build is a mess to tinker around but its the better engine. Lucasarts engine is better than both anyway. Shame its not open source.

>> No.6418716

>>6418706
Rumor says it's hacked together from reverse engineered Doom and Build code, and that's why they are not too keen about releasing the source.

>> No.6419549

>>6413979
it's a shame they're owned by WB now, the NOLF games, Blood and FEAR are some of the best shooters out there.

>> No.6419561

>>6416102
We're talking about a game using a lot of new, complex tech that only Carmack knew how to use at id, so I don't blame Romero with being unable to keep up. Supposedly everyone at id was sitting around waiting for Carmack to get the tech ready, so I imagine Romero was doing something during the day. Unlike previous years he wasn't busy licensing the Doom engine off to anyone or supervising expansion packs.

>> No.6419579

>>6414323
Ken made a ton of cool stuff after build, just not commercial games.

Evaldraw, which is a very cool tool for prototyping graphics and voxel code, I think he made a new version for modern gpu shaders as well.
Multiple voxel renderers including voxlap which he released the source for.
Build 2 which he recently released, and a whole lot more in his website. You should check it out.

Both carmack and silverman are great programmers but I think the main difference between them and normal people is an insane work ethic that allows them to finish much more work than most people. I'm a pretty decent programmer, but I have nowhere near the work ethic of either of these legends.

>> No.6419586

>>6410964
Abrash is pretty legendary, but he never made famous game engines like doom on his own. He was pretty instrumental in getting quake to run in real time though, some of his code in there is fucking savant level genius (the trick for calculating UV coords with a lookup table affine in the span rendering code, just how the fuck do you come up with that)

>> No.6419591

>>6419586
Typo: in the affine span rendering code

>> No.6419606

>>6418676
Fuck no. Ken has said in interviewx hes not really even into games anymore nor does he play them and Carmack has better things to do like launch fucking rockets or drive ferraris.

I could see Romero lurking here honestly, he's a more down to earth bro type and is active online and usually lurks places like doomworld, etc... but who knows? Its doubtful though even in his case.

>> No.6419613

>>6419606
Carmack sold his Ferraris and no longer works in rockets

>> No.6419619

>>6419549
Is FEAR really that good? I have had the 360 version laying around for years and never even played it beyond maybe part of the first level. I think The Ring looking bitch on the cover was off putting to me for some reason, looks jank. I fully admit I have not given the game a chance as of yet but I hear people gush over it all the time. I am not into Resident Evil type horror shooters though, is it like that?

>> No.6419625

>>6419613
Well then working on VR shit.

Come on, he's not fucking here, that's for sure.

>> No.6419656

>>6419619
I recommend playing it on PC, I think both console versions have problems. The creepy girl shows a J-Horror influence and there are a lot of scripted scares. It's scarier than Half-Life although FEAR is actually a fast-paced action game where you largely fight soldiers. It also has a bullet time mechanic like Max Payne. Level design and AI are very good.

>> No.6420004

>>6419656
Sweet, I will give it a go then. I will probably just throw the 360 version on and actually give it a little attention to see if I am going to like it and if I do I will go to PC.

>> No.6420025

>>6407634
>Ken Silverman
Maybe it's simply because I know more about Carmack than Silverman, but Carmack actually provided innovations that Silverman and others utilized.

>> No.6420041

Carmack recently said if he could do it all over again, he would hold off on Quake for a year and instead build a "Doom Next" engine instead.
Would be interesting to see if Carmack's next gen sector-based engine would run circles around Silverman's engine.

>> No.6420052

>>6420041
Quake is perfect though.

>> No.6420120

>>6420052
I wish it had more lore though... not asking for some cinematic driven epic but maybe a little more clear on exactly what the fuck is going on besides you're going through a slipgate to other dimensions which all look similar to destroy something called Quake that you don't even know what is

>> No.6420206

>>6418329
>longevity and ease-of-use/modding
Which wasn't actually Carmack's doing, since the actual tools Romero and gang worked with weren't released until way after the end of the century, while Build tools were there from day one. It's just a case of a community being far more determined to suffer through the hard work in order to push accessibility above all else. It wasn't originally like it is today.

>> No.6420210

>>6420025
True, but Carmack still considered and respected Silverman as a competitor. This quote is from 1999:

>"I certainly respect the abilities of my primary competitors. Back in the DOOM days, Ken Silverman was extremely impressive, and today Tim Sweeney is producing much of value."

>> No.6420227

>>6418716
That's grade A bullshit to anybody who actually worked with the engine itself, and most of all the Jedi Engine was coded around the same time as Build with Dark Forces releasing a whopping 7 months before the first proper Build game ever got to the shelves.

>> No.6420279

>>6420227
Why won't they release it then you shill? Do you work for LucasArts? I see no other reason you would be so disingenuous.

>> No.6420330

>>6420025
Silverman built what was far and away the most impressive voxel engine of the early 2000s.

>> No.6420832

>>6408875
It ran good on a 486, I had a lot more fun with multiplayed duke 3d than quake. I find the levels better and you have better interaction with descruction, switches and scripted sections. The graphics are much brighter and varied.

Ken moved on because I think he thought being a game programmer didn't have much of a future. Its not like the quake 3 engine really influenced the ps2 generation even though new call of duty games sort of use it.

>> No.6420963

>>6420832
>Ken moved on because I think he thought being a game programmer didn't have much of a future.
He did only because no game company wants his voxel technology. He was working on his voxel technology for more than a decade but nobody cared about it. So no he works in a hologram company instead where his voxel knowledge is useful.

I guess it's the same case with John Carmack and his VR obsession, except VR is hot garbage while voxel could actually revolutionize the game industry.

>> No.6421421

>>6420963
Remember the VR thing in 90s?
They were so ambitious back then.
The hardware couldn't meet the expectation yet.

>> No.6421480

>>6420052
doom is more fun to play than quake

>> No.6421502

Carmack, this isn't even a debate.

Silverman was a midwit autist who completed his engine through brute force. It's still CRAZY impressive that he fucking did it and he deserves a fuckton of praise, but the fact is that he got by on pure work ethic and persistence. There's a whole article where the Monolith liaison who had to work with him when Build was licenced for Blood tried to talk to him about adding features to the engine/streamlining poorly programmed parts and it was like pulling teeth because Silverman never had a fucking clue what he was talking about. There's a reason he hasn't been relevant since 1997.

>> No.6421646

>>6420227
>That's grade A bullshit to anybody who actually worked with the engine itself
Behold, a LucasArts veteran has honored us with his presence!
top kek

>> No.6421919

>>6420330
Don't forget Kyle Freeman's Comanche series

>> No.6421938

>>6407634
whatever happened to Silverman?
I think he was a genius and back then superior but he is long gone and vanished while Carmek is still around so what happened?

>> No.6421972

>>6407634
you know Ken wins because of the music he created
>https://youtu.be/9ZNR4PPmyp0
>https://youtu.be/XbSHkB3vF4w

those will forever be masterpieces no one can top

>> No.6421980

Ken sent me

>> No.6422183

>>6421502
Or maybe Ken's team wasnt as helpful as carmack's. Just because ken has no leadership, presentation, and documentation experience doesnt mean he was a bad programmer. He was 19 when he created the build engine while carmack was 23 when he made doom.

The reason he hasnt be relevant? Its because he only made voxel engines. Voxel stopped being popular after video cards got introduced. Video cards dont handle voxel.

>>6421938
Now works in a hologram company, developing the holographic devices and voxel graphics for use in biomedical researches and engineering, as well as the projector you see in some hatsune miku or eazy e concert. Basically, he's a programmer specialized in voxels, but since modern games rarely use voxel anymore, he's using it to make anime real instead.

>> No.6422226

>>6422183
Ken didn't have a team.

>> No.6422228

>>6421980
What's the ol' shylock up to these days?>>6421502
I think you're talking about the developer rants in Blood about Ken and 3D Realms posted here? >>6408306

>> No.6422237
File: 389 KB, 720x459, content_dam_lfw_online_articles_2015_july_voxiebox.png [View same] [iqdb] [saucenao] [google]
6422237

>>6422183
>>6421938
Check out this presentation he did for the hologram game thing he was working on called VoxieBox. He is a total sperg who had to take this opportunity to flex about Duke Nukem 3D and the Build engine even though it's pretty irrelevant to the presentation itself, but the product looks impressive as fuck, I don't know why this thing never took off

>> No.6422240

>>6422237
Fuck, forgot link:
https://www.youtube.com/watch?v=-q2zxXrtz1c

>> No.6422418

>>6422183


Nobody doubts that Ken was way beyond any rasterizing enthusiast in raw skill.

>> No.6422429

Ken seems pretty depressed after having lost the bleeding edge and thus 3D-fame in '95 / '96.
Hope he doesn't turn into a SilverBareBack yearning for a mount by "Big" John.

>> No.6422713

>>6421972
Hell yeah quicksong

>> No.6422720

>>6421919
That was a heightmap raytracer, lots of people count that as "voxels", but it's much much simpler and less sophisticated than Voxlap.

>> No.6422802

>>6422720
Interesting. I remember when playing Comanche though, if you landed your chopper you could literally see the individual voxels/squares. So this was an optical illusion and not 'real' voxels?

>> No.6422824

>>6422802
It's basically an illusion. The data for that map is a 2d grid, and each cell in the grid contains a single number, which is the height. With clever programming and good math, you can draw it out so each square is visible and looks cubic.

Voxlap is a 2d grid as well, but each grid cell contains a list of voxel runs, including empty space, so you can have any level of geometric complexity.

>> No.6422851

>>6422802
>>6422824
I guess I should be fair and say that a 2d heightmap is a fine way to encode volumetric data, as long as you don't need overhangs, caves, buildings, etc

>> No.6423090

>>6422183
>He was 19 when he created the build engine
He was 19 when he finished, it took him years of trial and error.

>> No.6423445
File: 14 KB, 550x550, ken_sent_me.jpg [View same] [iqdb] [saucenao] [google]
6423445

>>6421980
>>6422226
Ken is da pimp.

>> No.6424308
File: 64 KB, 900x450, download.jpg [View same] [iqdb] [saucenao] [google]
6424308

>>6422240
>we don't want people to see the inside just yet

>> No.6424354

>>6410201
>>6410678
For what it's worth, there were a group of Hungarian scientists referred to as 'The Martians', and a number of them were of Jewish ancestry.

>> No.6424591

>>6409364
voxlap was full 3d with complete destructability

also ken was working on a quake level engine, but got pulled back to help teams that were still using the build engine. by the time the build games were done, quake was already released (and possibly quake 2?), so he felt he was very behind and iirc went back to university.

it's likely he would have had something very impressive for the time as well if allowed to complete it.

>> No.6424594

>>6410861
what exactly is a mess about it? It reads fine to me, you just cant skim over it like most people expect to, thinking they're "reading" it.

>> No.6424798

>>6424591

Exactly - he just missed some time and the gap was closing too fast.

Maybe if he swallowed his youthful pride he could have joined id and aided them in development. I guess he was too afraid that Carmack was "better" and mentally pulled out. A total waste - Abrash, Carmack & Silverman would have been an awesome treat.

>> No.6425006

>>6424594
I don't know why they are so butthurt about it either. But it might just be that they are the kind of people who find hungarian notation beautiful...

>> No.6425725

>>6424798
He doesn't seem like the kind of person to work on a team. Maybe nowadays, but a probably mildly autistic guy in his teens? Programmers already tend to be loners, and people like that always work alone.

>> No.6425812

>>6416090
>I think many of the devs that worked with Ken became hostile because they couldn't keep up with him - both positively and negatively speaking(talent and chaos wise).

Nah, the build code was and still is a complete fucking mess, he was 18 yrs old and inexperienced at the time and it shows

>> No.6425881

>>6409225
>LARPer

>> No.6426387

>>6421972
What format is the music in? Anything standard, or did Ken make something unique?

>> No.6426432
File: 11 KB, 316x160, nasir.jpg [View same] [iqdb] [saucenao] [google]
6426432

>>6407634
NASIR

>> No.6426472 [DELETED] 

>>6416752
do you have delicious snack to back up that hot delishness?

>> No.6426521

>>6426432
NASIR is impressive, but his games are buggy as shit.
Secret of Mana is more glitch then game.

>> No.6426584

>>6426521
Really? I know half the stuff in FF1 doesn't work, but SoM? Care to post some examples? Because I never noticed anything wrong with it.

>> No.6426695

>>6410219
>>6410225
Based on what?

>>6410224
When you see a line skip just to separate two sentences, it's evident they came from there. If it's to separate two paragraphs, it's fine.
What's funnier is how pointing it out infuriates them, because they're so used to needing to do this over there.

>> No.6426714

>>6414498
Why couldn't they have been based rural redneck coders with diabetes? Imagine how many more NASCAR games we could have gotten, fellow centipedes?
We truly are an oppressed peoples, I reckon.

>> No.6426718

>>6415607
use the futaba theme, then. Nobody's stopping you. Hell, you could do it right now.

>> No.6426728

>>6416374
Why are you indulging his schizophrenic bullshit? You're only making his problems worse.

>> No.6426747

>>6426714
Oof! Upboated and gilded, my friend!

>> No.6426768
File: 1.72 MB, 2628x1778, Screen Shot 2020-05-13 at 3.39.16 PM.png [View same] [iqdb] [saucenao] [google]
6426768

got some inspiration from silverman and voxlap to make a quick voxel demo.

only 4DOF, but supports arbitrary columns of voxels, albeit slowly.

stole the graphics from comanche, span buffers are kinda tricky so it has some clipping issues

>> No.6426776

>>6426747
>*impotent yokel noises*
Very nice.

>> No.6426792
File: 1.80 MB, 2584x1738, Screen Shot 2020-05-13 at 3.49.17 PM.png [View same] [iqdb] [saucenao] [google]
6426792

>>6426768
also, i didnt put in the effort to get the grid intersections exactly right, so no perfect cube voxels.. maybe i'll do that next

>> No.6428135

>>6426768

Awesome!

>> No.6428215

Carmack would snap Silverman in half, come on.

>> No.6429056

>>6411054
>Carmack was 22 when he wrote Wolf3D and started work on Doom, completing it at 23.

I'm 22 close to 23, and your comment made me reflect for a moment about my accomplishments...

Nothing.

>> No.6429174

>>6418527
They absolutely loved that guy!

>> No.6429191

>>6420004
I loved FEAR, hope you do as well :) unfortunately, I haven't played its successors, might give 'em a go.

>> No.6429270

>>6429191
Been so busy with work I haven't had a chance to play it yet but I plan on it this weekend, since I'll be off. Is there alot of action? I have been binging on Doom and Dusk lately and I don't know if it'll hold my attention if it isn't

>> No.6429386

>>6419606
>I could see Romero lurking here honestly
Maybe he really did make those Daikatana threads after all

>> No.6430929

>>6426792
>>6426768
Do Comanche and Silverman's work use the same set of principles for voxel rendtition?

>> No.6430954

>>6430929
Comanche has a 2d grid that describes the height and color of each point of the terrain.

Voxlap has a 2d grid, where each slot in the grid contains a list of runs, where each contains a color and the length of the run. I assume there's a placeholder color, probably full transparent, that represents empty space. Each column can be up to 256 high if I remember correctly, but this could be changed in the source code with modern hardware.

I took the Comanche data and converted it to a voxlap style data structure with three runs per column, the terrain, a bunch of empty space, then the same terrain shifted over a bit, and vertically mirrored.

>> No.6430961

>>6430929
Oh, I just realized you asked about rendition and not the storage. Comanche uses a simple raycasting algorithm, and my program does the same thing extended to handle multiple runs per column. Voxlap is much more complicated because of the 6DOF support. Think about how basic raycasting would work if you looked straight at the ground. Not very well. I actually haven't written a program like voxlap so I couldn't say I understand it that well, but there's a couple of papers describing a similar technique if you're interested.

>> No.6431524
File: 1.68 MB, 3264x2448, zfight.jpg [View same] [iqdb] [saucenao] [google]
6431524

>>6430961
I'm by no means knowledgeable in the field, I was just wondering how something running in 1992 still struggles on modern platforms. Obviously, the lack of acceleration is a major factor, but I have seen some impressive things done on modern CPU's (like a z-buffer renderer that basically ties with my GPU by using SIMD and clever tricks).

>> No.6431545

>>6429056
Make sure to tell /r9k/ about it and refrain from posting here.

>> No.6432095 [DELETED] 

>>6431524
It's written in javascript and totally unoptimized. If you remove the span buffer and just draw a single column of voxels per grid cell it'll hit 60fps. The same thing in C with proper integer raycasting would probably hit several hundred.

>> No.6432107

>>6431524
It's a quick demo I threw together in javascript in about an hour, so it's totally unoptimized. If you remove the span buffer and just draw a single column of voxels per grid cell it'll hit 60fps. The same thing in C with proper integer raycasting would probably hit several hundred regardless of how many voxel columns you draw.

>> No.6432329

>>6424591
>by the time the build games were done, quake was already released (and possibly quake 2?), so he felt he was very behind and iirc went back to university

too bad by that time gpu cards became a thing and they invalidated making engines as they knew it. from then it was all about putting triangles to screen and assigning textures to them. the focus moved on from making epic 3d algorithms to designing engines in such way to utilize the hardware power the most to empower the rest of game engine and not get in the way of it (something Ken was awfully incompetent at)

>> No.6432346

>>6432329
which is why he kept working on his cool graphics algorithms like voxlap, which was possibly a better outcome than yet another hardware accelerated triangle renderer.

>> No.6433690
File: 788 KB, 2628x1760, cubes.png [View same] [iqdb] [saucenao] [google]
6433690

got cube voxels kinda working

>> No.6433692

>>6407683
He sold it to Facebook instead

>> No.6435370

>>6433690

Awesome!

>> No.6437556

>>6407683
Facebook poses a bigger threat to the western world than tencent.

>> No.6437587

>>6433690
Are linear or cubic filters feasible for voxels like they are with bitmaps for smooth geometry?

>> No.6438335

>>6437587
Yes, I like the cubes though

>> No.6438339

>>6407634
Not much competition. While Silverman's Build does blow away Doom, Carmack went on to do a whole fuckton of other things. Carmack is a genius when it comes to coding. Dude isn't human.

>> No.6438342

>>6407634
Tim Sweeney.

>> No.6438345

>>6408740
I don't think the term sellout is fair. The whole reason he made the Unreal game and packaged it with the builder software was to lay the groundwork for a generation of programmers to use his engine. He never set out to make video games, his vision was to make a standardized toolset for all games to be made. And frankly, Epic business practices aside, I think it was one of the best things to have happened to video games.

>> No.6438363

>>6438339
You probably don't know much of Silverman's history if you don't think his programming skills aren't incredible, and don't know about his projects after build. I'd say the biggest difference is that Ken has also been a lone programmer working on whatever projects he wants to, whereas Carmack learned how to work with teams.

>> No.6438374

>>6410964
Unreal engine software transform code is a copy of Mike Abrash fast transform routine published in one of his books.
Don't need to believe me, just look at the source code. There's a comment in the function stating exactly that.

Mike Abrash is a genius.

>> No.6438396

>>6438363
The team, especially in the early days durig development of Doom and Quake didn't really do shit to help Carmack on the engine, as it was above their skill levels. The whole drama with Romero during development of Quake was due to this - Carmack and others accused him of not contributing and fucking off all day, while his defense was that they couldn't do anything until Carmack finished the engine lest they end up with a bunch of work that is useless with the finished engine.

>> No.6438421

>>6438396
The general work flow for Carmack during Quake was:
Carmack writes a portion of the engine code
It's passed on to Abrash who optimizes it

I'm sure it sucked if you were an artist or level designer, but Carmack has a history of splitting work with another programmer.

>> No.6438429

>>6438421
I forgot about Abrash.. I mean yeah in that regard I suppose he had help that Silverman didn't have but at the same time I don't think Carmack would've been unable to do it on his own either, it just would've took him longer

>> No.6438441

>>6438396
I mean that Carmack code is at least generally designed with other people using it in mind. Silverman's code looks like it's written for his use alone, even if he does release the source code.

>> No.6438517

>>6438441
Oh now that is certainly true - Silverman's code is an absolute mess and he essentially brute forced his way through and bolted on additional shit as he went - just looking at it tells you that he never intended for anyone to see it but him

>> No.6438575

>>6438517
to be fair gamedev in general is just bruteforcing it

especially when you are trying to come up with new algorithms

>> No.6438769

Carmack, a literal rocket scientist created the most efficient ways to render polygons EVER. Fucking Half-life and the Call of Dookie shit the kids play today wouldn't exist if it wasn't for QUAKE.

>> No.6439164

>>6438769
Quake was great for its time yeah, but calling it "the most efficient way to render polygons ever" just shows you have no idea what you're talking about.
Quake has multiple polygon renderers because different things are more efficient depending on the use case. And what's most efficient also depends on the hardware.

>> No.6439326

>>6438769
Also, gpus don't render polygons in the same way as quake

>> No.6439518

>>6438769
>Fucking Half-life and the Call of Dookie shit the kids play today wouldn't exist if it wasn't for QUAKE.
well no.. they wouldn't.. because hl uses a modified quake1 engine, and cod1 runs on the quake3 engine

>> No.6439890

>>6439518
Also, unlike games like return to castle wolfenstein, cod runs on a heavily modified quake 3 engine that lets it support an incredible draw distance, high enemy count, and more advanced gun physics. Doesnt matter if it runs on idtech or not, its almost unrecognizable and far better than all the games actually built using that engine. I bet they only used idtech because its supported by modern video cards.

>> No.6441636

>>6407676
RE2 on the N64

>> No.6442172
File: 2.87 MB, 640x360, voxel.webm [View same] [iqdb] [saucenao] [google]
6442172

>>6437587
now that I think about it, doing a filter on this data (each column can have multiple voxel chunks) would be quite a bit trickier than a basic heightmap

>> No.6443507

>>6442172
reminds me of that one game i've played that uses a voxel landscape
https://youtu.be/ld-7j6d7Yhs?t=20
always amazed me how it looked, everything was so smooth compared to it's contemporaries

>> No.6443551

>>6438429

Most likely. Abrash had more experience than Carmack and was able to compliment him "perfectly". I read somewhere that he also found the Pentium bug before it was announced and had an intel engineer come to the id office to pin point the error. Very interesting times back then.

>> No.6443559

>>6438517

And where's the problem? He accomplished what he intended to.

>> No.6443638

>>6441636
Amazing compression but the port sucked. WWE No Mercy is more impressive and fun.

>> No.6445731

>>6407634
carmack, no competition.

john carmacks name is recognized throughout industries world wide in technology, from both past and present.

the same can't be said for silverman sadly.

>> No.6445742

>>6445731
>popular good
>unpopular bad

>> No.6445752

>>6445731
Surprisingly. He must be such a fucking sperg that not even Jewish nepotism could get him the renown Carmack earned

>> No.6445754

>>6445742
no, its usually just the tell tale signs that if you're good at something it will dominate the rest of your competition and live on as it has provided you more resources to continue improving.

or is that weawwy just too much fow yow swall wittle bwain to compwehend?

>> No.6445757

>>6407634
Tim Sweeney probably.

>> No.6445758

>>6445752
exactly.

>> No.6445761

>>6445754
>ken isn't good at programming

what did he mean by this

>> No.6445764

>>6407634
Unless you're talkin' retro-for-retro there's no way in hell this is a fair tossup.

>> No.6445791

>>6410080
Quake 3 was better atremdering curved surfaces. Did everything else worse.
>>6410095
What specs? Because the only issues I've ever had have been running UE1 games on modern machines. Not UT or Unreal - they've always ran fine - but Klingon Honor Guard and the like.
Hell Unreal used to run on our old 233mhz piece of shit. Not well, but playably.

>> No.6445802

>>6445761
you know that green text is used for quotes? or are you new?

>> No.6446794

>>6445802
I've been here since 07, newfag