Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Steam: Half-Life 2 Hardware Survey (2004)

web.archive.org

Post reply on HN
Print view View original

151–160 of 190 posts

Re: Steam: Half-Life 2 Hardware Survey (2004)

#151

Post by amlib »

Zardoz84 wrote: post #45
skocznymroczny wrote: post #32

I remember running Half Life 2 on some card way below the recommended spec. There was some commandline switch to run the game in DX7 or DX8 mode, turned everything to lowest and resolution like 640x480. Looked worse than Half Life 1, but at least was somewhat playable.

Some people managed to run Doom 3 on Voodoo 2/3 cards, applying some patches and resizing the game textures, etc... I managed to run Doom 3 on a Creative GeForce 2 MX DDR using the patches for Voodoo 2 and doing the same trick of resizing textures. The game was playable, except for the boss guarding the soul cube.

I was one of the poor bastards playing Half-Life 2 on a Voodoo 3 around that time... stuttered like crazy, required setting everything as low as possible and then some with console commands... ran at an admirable ~20fps, but I was used to worse on my n64, so I just sucked it up lol.

Looking at some screenshots, it didn't look all that bad... https://imgur.com/a/cwBHywJ

Re: Steam: Half-Life 2 Hardware Survey (2004)

#152

Post by mekkkkkk »

tomc1985 wrote: post #139
mekkkkkk wrote: post #65

Earlier quoted context omitted.

Regarding the first part, I'm pretty sure that Steam had an offline mode from the start. Perhaps it required an initial connection to set up an account, though. I was lucky to be in an area that had very early access to DSL, so it wasn't a problem for me. The "always on" features was actually pretty awesome and novel if you had a permanent connection. In general I've been impressed by Steam from the start. On the oth…

> If anything it's now more confusing than ever to keep track of which launcher has which game and which friends had been added where. I've been using Playnite to consolidate my games list across launchers. It works pretty well once it's set up, except for titles where the launcher you bought it on launches another launcher

> except for titles where the launcher you bought it on launches another launcher

That's just comedy. I know this happens for a lot of Ubisoft games in my Steam library. I guess they want to keep the cake and eat it to. Exposure on Steam, walled garden in Connect.

I'll definitely check out Playnite. Thanks for the tip!

Re: Steam: Half-Life 2 Hardware Survey (2004)

#153

Post by paulmd »

_fat_santa wrote: post #113
tikkabhuna wrote: post #34

Earlier quoted context omitted.

The Crysis 1 -> 2 experience felt so disappointing. I loved how Crysis 1 there were no limits to the weapon attachments or switching suit functions. I remember Crysis 1 with the silenced pistols with laser pointers you would 100% hit where the laser was pointing. I'd go full cheese and sneak up to the enemy, de-cloak, 1 shot dead, re-cloak. All to push a story which started with an off screen killing of Nomad.

One of my fondest memories around Crysis is the whole "but can it run Crysis" unofficial benchmark. I remember everyone looking at the game going "yeah my machine can't run this". Ended up the game ran pretty well even on my potato machine back then, just had to keep all the graphics on low. I'm kind of tempted to buy the remastered version and see how much better it looks on my Xbox Series S like a decade later

Unfortunately, Crysis 1 is seriously single-thread bottlenecked. It was designed in anticipation of the "10 GHz Pentium 4" school of CPUs that never ended up existing, so it runs like trash even today with high-end hardware.

The remaster does not fix this and actually is a step down in visuals in some respects, due to the ever popular "we remastered from a shitty port that never worked right in the first place". Halo also comes to mind, with the Halo PC version having completely broken shaders and much different weapons balance from the xbox version. In the case of crysis I believe it was the other way around, the game was remastered from the lower-fidelity Xbox version... so it actually is visually worse in some areas.

https://gamerant.com/crysis-remastered-review/

https://www.youtube.com/watch?v=l_Az4-2o9AI

Re: Steam: Half-Life 2 Hardware Survey (2004)

#154

Post by Vecr »

kevingadd wrote: post #122
ChuckNorris89 wrote: post #119

Earlier quoted context omitted.

>Fun fact but for years updating GPUs led to a decrease rather than an increase in framerate. Sounds more like a specific technical issue with an specific engine and drivers that lacked optimization rather than an axiom that can generalize games and GPUs for several years. If you have a source for this please share it. Also, IIRC, back then the game physics were tied to the framerate, so having too high framerate mad…

In the modern era we still have problems of this sort - higher core counts can cause games to perform worse due to problems like a game suddenly having 96+ threadpool threads competing for small amounts of work, often belonging to the driver. It's still the case that a high core count Ryzen will have worse performance in some games (or problems like microstuttering) that you won't experience on a slower, lower core c…

In that case you might want to mask off some of the cores/hyperthreads, or even do core pinning. That's very common in applications where latency is important.

Re: Steam: Half-Life 2 Hardware Survey (2004)

#156

Post by ridgered4 »

RF_Savage wrote: post #145
filmor wrote: post #70

Earlier quoted context omitted.

The GeForce 4 MX (first in the list) was a rebranded GeForce 2, so just T&L and no shaders. The "real" GeForce 4 (second in the list) supported shaders just fine, as did the GeForce 3, so it got the pretty water.

And yet, my box with a 50eur GeForce4 MX440 with 64mb VRAM ran Half-Life 2 on launch. AMD Duron 1800+ and 256mb of 333MHz DDR and the damn thing ran HL2! A machine with a 50eur CPU and 50eur GPU ran it! They must have optimized it well. And that machine is included in that 2004 Steam hardware survey. :)

There was a very different attitude to budget offerings from nvidia back then, and despite a lot of complaints about them I'm not sure the new way is better.

Nvidia seemed to like to take their old hardware design and crank it up a bit, slap a new version on it and sell it as a budget offering. No new features, but it did a respectable job of playing older games which is what I think most people want out of a cheapo card. Later offerings usually featured super cut down cards that had more of the new features but not really enough power to actually turn them on and get playable framerates. And that is how things have sort of gone since then.

Like you said, the GF4 MX440 was actually a decent performing card. Sure, you were stuck with DX7 rendering path but it did a good job with it.

Re: Steam: Half-Life 2 Hardware Survey (2004)

#158

Post by ridgered4 »

hsbauauvhabzb wrote: post #129
mekkkkkk wrote: post #41

HL2, Portal, CS:S. Valve really hit it out of the park in those days. I wonder how much of Steams success can be traced back to it. Just the Valve catalog alone carried it for me and my friends.

Loaded up the Ubisoft app recently. Pure trash, apps / syncing wouldn’t work till I accepted all cookies, things wouldn’t download in the background while playing single player, etc.

I tried playing an old Ubisoft title recently, Farcry Blood Dragon. But it wants to connect to a long dead ubisoft server EVERY time I his Esc to go to the menu and takes 30 seconds to time out. Every time! I decided I just won't buy anymore Ubisoft games after that.

Re: Steam: Half-Life 2 Hardware Survey (2004)

#159

Post by ridgered4 »

BlueTemplar wrote: post #107
AnIdiotOnTheNet wrote: post #97

Earlier quoted context omitted.

A lot of us predicted horrible things would come from Steam. Ads everywhere, forced updates, forced removal of games for all kinds of bullshit reasons, etc. Basically all the same arguments you get today about digital ownership and platforms. Fortunately for us, Valve has actually been a pretty good gaming-community citizen with Steam so we can happily say we were wrong.

Forced updates did happen (only way to play the game if Steam was aware there was one available - and God forbid you had a slow/flaky connection - no gaming for you today, not even through Offline Mode !). Combined in at least one instance with the new version not being able to load the old multi-hours save, and no rollbacks, so hopefully you manually backed up the whole game ! (At least the devs fixed it in a patch…

I'd say if I have one big complaint about Steam, it's that it's update control needs work. I don't know if it's intentional or not but rolling back to an old version is...impossible? I'm vaguely aware there is an option to maintain the version you already have but I believe it's quite buried.

I've had updates completely change game balance, making character load outs I'd grinded up not work the way they did before. I've also had new linux versions break compatibility. This is understandable in multiplayer games, but I was playing these games single player.

Re: Steam: Half-Life 2 Hardware Survey (2004)

#160

Post by BearOso »

jimbob45 wrote: post #116
epolanski wrote: post #112

Fun fact but for years updating GPUs led to a decrease rather than an increase in framerate. Basically with a solid GeForce 4 series card you could already cap the 100 fps framerate on Half-Life, but update to a much more powerful 9600 pro and framerate would start getting much more lower. The Quake 1 engine would be so old that getting better GPUs would make it run worse. Also, another fun fact about HL is that scal…

On that note, I still regularly get extremely bad FPS on Quake-based games today with my RX 580. Only renaming the executable seems to solve the issue. https://techreport.com/review/3089/how-atis-drivers-optimize...

It's amazing how many workarounds never get removed from big codebases. For example, look at Chrome/Chromium's about:gpu page. Some driver workarounds are still applied for bug reports that haven't been looked at for a decade.
Post reply on HN