Doom - Crysis of it's time?

Was Doom a "Crysis" of it's time? I have read forums and watched gameplay videos on 386/486, and tried my own 486.

Doom requires minimum 386+4mb, recommends 486+8mb, or so they say. Isn't this "requirement" just very wrong/incomplete?

I mean 386 clock speed ranged from 12 to 40mhz, one 386 can be 3 times faster than another one! But even fastest one, 386dx-40mhz with 4mb ram is barely playable. Now, I know that acceptable fps changes from person to person, but I think most claims are affected by memories of the past and memories aren't always the way a person remembers them. I am not a "hd+minimum60fps" requirer, but I also know when the game becomes unplayable. For me, anything below 0.04Mpixels(200x200) and 12fps is unplayable/unacceptable for a game like doom. Seriously, people have played it at much lower specs than what I mentioned. "Postage stamp screen size", slideshow 3-4fps, etc... From what I understood most 386 simply is not capable of 0.04Mpxl@12fps, neither is the slowest 486 (486-16mhz), I am not sure about 486-20mhz. Now, talk about 486-25mhz.

I have a 486-25mhz 4mb-ram laptop. "doom -timedemo demo3" yields 8.1fps on it on screensize 9 (fully 320x200, fullscreen, hud is not visible). I can barely bring it to just over 9fps while still staying above 0.04Mpxl, though my system has no L2. So I predict a "good 486-20mhz system" should suffice. So, from my experience, I would say a "good" 386-40mhz OR a "good" 486-20mhz with 4mb-ram is the absolute minimum for a playable doom experience.

On the other hand, this video https://www.youtube.com/watch?v=8iH07x_Q6mQ

Damn, 486dx2-66mhz and 32mb-ram, 21fps. What the hell does it take to run this thing on 35fps ?!

Sidenote: just for curiosity I ran doom benchmark on my 2007 laptop with core2duo 2ghz and infinite ram in real ms-dos 6.22:     2134 gametics in 478 realtics, that is 156fps!!! (156fps would be achievable in game if it did not restrict itself to 35fps), also it has 1280x800 resolution, a perfect x4 scale-up of 320x200, though the soundcard on these things is not soundblaster compatible, so say hello to pc speaker.

Definitely. IIRC Doom didn't run at 35 FPS with maxed-out settings until Pentiums arrived, or if you used one of those AMD 486s that ran at a 80-100 MHz clock rate.

There's a pretty comprehensive list of systems tested here: https://www.complang.tuwien.ac.at/misc/doombench.html

The reason for sticking to one benchmark is simple: It allows for direct comparison. I don't know why specifically Demo3 was chosen over the others, but since the content is exactly the same, and (aside from speed) the demo will play out the same way every time, it gives an unchanging base to compare against.

You can't just look at the cpu speed here. The video card also played a big role in how fast Doom ran. The game doesn't use any hardware acceleration as we know it today, but it does benefit from a faster graphics card since it sends a lot of pixel data to the card. Same with memory and cache. And not all 386 designs are equally fast.

Pentiums were released in march 1993, but weren't in common use for gaming. A P60 should run Doom comfortably. A 486 DX2/66 MHz (August 1992) should also run Doom fairly well when paired with a good graphics card etc. At least by the standards of that era.

Intel had a 486 DX/4 100MHz (actually tripled 33MHz). Other manufacturers had even faster 486-like designs.

The instruction pipelining on the 486 was far superior to the 386 which had a design flaw that meant it stalled more often than desired. That was a significant reason the 486 was faster even at the same nominal clock rate. If you like this stuff, you'll love Fabien's book: http://fabiensanglard.net/gebbdoom/

https://en.wikipedia.org/wiki/Origin_Systems

The old joke used to be that when Origin released a game, you'd have to play with your config.sys/autoexec.bat and/or upgrade your system to get it running.

Like QUAKE for 486, QUAKE 2 for slow pentium, QUAKE 3 for pentium 1, 2, etc :)

EDIT:

ha, I've just remebered unreal ram recomandation: 128, in time when people got 16 MB, lol.

I'd say software Quake was much more of a "Crysis" than Doom was. Without GL, Quake was not playable at all on most computers, and all those high resolutions were completely useless for 2-3 years after Quake came out because the CPUs available were just not powerful enough. GLQuake was clunky, ugly, and generally shitty compared to modern Quake ports and looked a whole lot worse than DOS Quake, but it didn't matter because it was playable on normal people's machines and DOS Quake was not. They called the Voodoo and friends accelerators back then for a reason, they made games fast, not pretty.

I remember very well like Quake run on k6 - that was first time when I could see 800x600 in smooth, but 1024x768 was laggy. Quake on slow pentium (only in DOS) was playable in one of 320xsomething. On windows with even 32 megs it was terrible. First GLQuakes were only good to see super effects like transparent water/windows and filtered texture (what was insane great these times).

EDIT: pentium III and than much better for gamers celeron was a great improve to play all quakes. Until DOOM 3 :D

2 hours ago, Woolie Wool said:

I'd say software Quake was much more of a "Crysis" than Doom was. Without GL, Quake was not playable at all on most computers, and all those high resolutions were completely useless for 2-3 years after Quake came out because the CPUs available were just not powerful enough. GLQuake was clunky, ugly, and generally shitty compared to modern Quake ports and looked a whole lot worse than DOS Quake, but it didn't matter because it was playable on normal people's machines and DOS Quake was not. They called the Voodoo and friends accelerators back then for a reason, they made games fast, not pretty.

Yeah, you pretty much needed a new rig to play Quake at a good framerate when it came out. I think Duke Nukem was more popular as you could still run it on older machines. As it happened we got Pentium Pros running Windows NT for the office around that time, and had no problems. I even managed to get the odd resolution of 360x400 running at a acceptable framerate.

GLQuake was a visible improvement on DOS at the time in terms of resolution, colour depth and filtering, and felt like you were playing on a Playstation. Frame rate on the other hand wasn't notably faster in my experience. Places in Quake where my old P90 slowed down, still slowed down with a Voodoo card installed. I guess the CPU, motherboard, etc remained a bottleneck for some things.

9 hours ago, Woolie Wool said:

I'd say software Quake was much more of a "Crysis" than Doom was. Without GL, Quake was not playable at all on most computers, and all those high resolutions were completely useless for 2-3 years after Quake came out because the CPUs available were just not powerful enough. GLQuake was clunky, ugly, and generally shitty compared to modern Quake ports and looked a whole lot worse than DOS Quake, but it didn't matter because it was playable on normal people's machines and DOS Quake was not. They called the Voodoo and friends accelerators back then for a reason, they made games fast, not pretty.

Did we play the same Quake games? GLQuake looked a ton better than what it did in software mode, and there was a lot of features people could turn on/off. Voodoo cards ran it fine in 640*480 with better fps than software 320*200. The defaults were conservative and designed around the original voodoo cards.

A lot of stuff was off by default, but with better hardware / less demaninding content it could be turned on for higher quality images. Texture filtering, high-res support, reflecting textures, dynamic shadows, transparent water. GLQuake had a lot of cool stuff: https://quake.fandom.com/wiki/GLQuake

I think a lot of stuff looked *worse* in the original glquake. Shadows lost their contrast and full brights stopped working. But it was nice to have sharp edges on brushes 

I don't know if I'd go so far to call it the "Crysis of its time". That would imply that a decade later, it would still run like hot garbage no matter the machine you tested it on.