• A Great friend to the HardForum with a great kid that he is trying to get a scholorship to continue his schooling. Please give hime a vote! Only 24 hours left! Thanks.
    If you have an VOTE FOR KEENAN!

CryEngine 2 Vs Source 07 Engine Vs Unreal 3 Engine

Which Engine is better?

  • Crysis's CryEngine 2.

    Votes: 160 46.4%
  • Half Life 2's Source 07 Engine.

    Votes: 93 27.0%
  • Unreal Tournament 3's Unreal 3 Engine.

    Votes: 49 14.2%
  • I dont really care.

    Votes: 43 12.5%

  • Total voters
    345
Ok, let me try and be more clear then.


Considering that the HL engine was a fork of the Quake engine, and a Valve developer states that Source is a fork of the HL engine, is it still not "clear to you" how there is a "development path forking directly from the Quake engine"?

That just proves my statement. These used parts of the Quake code to help test the Source engine during the beginning of its development, which were later removed. Thanks jimmy for helping me out, I knew you would come around. ;)

[EDIT] And just in case if people don't understand what I'm talking about, here's the line that comes after the quote Jimmy posted above, "Source was developed part-by-part from this fork onwards, slowly replacing GoldSrc in Valve's internal projects and explaining in part the reasons behind its unusually modular nature".
 
That's not part of the quote I posted. In any case it's not relevant to what I am asserting. I just wanted to establish that Source is indeed a fork of a fork of the Quake engine.

Now, regardless of whether they've removed all the Quake code or not, do you still fail to see that Source has a "development path forking directly from the Quake engine"?
 
That's not part of the quote I posted.

Yes it is, you quoted wikipedia which didn't include the full quote.

Now, regardless of whether they've removed all the Quake code or not, do you still fail to see that Source has a "development path forking directly from the Quake engine"?

Not at all. Just because they temporarily used portions of the GoldSource engine to test certain portions of the Source engine doesn't mean the "Source engine is ultimately based on the Quake engine" (which is what you originally posted).

I see you have tried to change your argument to "development path forking directly from the Quake engine" (I'm assuming because deep down you knew your first statement was quite incorrect). However this is still incorrect because Valve never used the GoldSource engine as a "developing path" for the Source engine, only as a tool to test the Source engine. Care to change your argument again?
 
I was quoting Valve developer Erik Johnson as indicated in my post, not wikipedia.

So despite the fact that Source is a fork of GoldSource (see the quote of Erik Johnson) and GoldSource is a fork of Quake, you are still unable to see any contiguous development path?
 
I was quoting Valve developer Erik Johnson as indicated in my post, not wikipedia.

And yet you don't want to finish (most relevant part of) the quote? Anything to fit your argument?

So despite the fact that Source is a fork of GoldSource (see the quote of Erik Johnson) and GoldSource is a fork of Quake, you are still unable to see any contiguous development path?

I believe if you read the whole quote you will be better informed (it's from the book that come with the super special edition of Half-Life 2). Perhaps you should change your argument again. Or maybe you can take some other quotes out of context to help yourself out.
 
I think you're misunderstanding me. I'm not denying that a considerable amount of code was replaced (perhaps even all of it, perhaps not), that's not relevant to my claim. I'm not sure how I can make that any clearer.

As far I can tell you think I am making two arguments: the "Source engine is ultimately based on the Quake engine if you trace it back far enough", and it has a "development path forking directly from the Quake engine".

Personally I meant these statements to have equivalent meanings, and I think it's a bit pedantic to insist otherwise, but if you're going to, I will retract the first statement for the sake of avoiding further argument (although I do believe it).

Can we at least agree that the Source engine has a "development path forking directly from the Quake engine" given that Source forked from GoldSource, and GoldSource forked from Quake? I'm curious to know your answer to this question having asked it several times already.
 
As far I can tell you think I am making two arguments: the "Source engine is ultimately based on the Quake engine if you trace it back far enough", and it has a "development path forking directly from the Quake engine".

Personally I meant these statements to have equivalent meanings

Can we at least agree that the Source engine has a "development path forking directly from the Quake engine" given that Source forked from GoldSource, and GoldSource forked from Quake?

Well since you believe those arguments are equal (which I don't see how, but alright) the answer to your question is quite easy, of course not. How can the Source engine be "ultimately based" on the Quake engine if it doesn't share any common code? What you're essentially saying is 4 is equal to 6.
 
As I mentioned in my previous post I retracted the "ultimately based" statement for the sake of limiting the breadth of the discussion (despite believing it to be true).

I'll ask again, can we agree that the Source engine had a "development path forking directly from the Quake engine" keeping in mind that it is a fork of a fork of the Quake engine?

If not, then I think we'll have to agree to disagree on this. I remain thoroughly unconvinced of your point of view, and I really don't know how to explain myself any better.
 
I'll ask again, can we agree that the Source engine had a "development path forking directly from the Quake engine" keeping in mind that it is a fork of a fork of the Quake engine?

Moofasa~ said:
Just because they temporarily used portions of the GoldSource engine to test certain portions of the Source engine doesn't mean the "Source engine is ultimately based on the Quake engine".

Replace "is ultimately based on the Quake engine" with " had a development path forking directly from the Quake engine" and you have my answer.

used as a testing tool != had a development path from

Valve could have used any engine (well engines they had the source code to) to test portions of the Source engine. They could have used an old Unreal engine, but that still wouldn't make the Source engine "ultimately based" on the Unreal engine. Obviously it was easiest to use their old engine, which happen to be a heavily modified version of the original Quake engine. You can believe whatever you want, but I don't think it gets any more clear than this.
 
You still haven't answered my question: Do you agree that the source engine has a "development path forking directly from the Quake engine"? A yes or no will suffice.
 
You still haven't answered my question: Do you agree that the source engine has a "development path forking directly from the Quake engine"? A yes or no will suffice.

Have you read any of my posts? :confused:

Moofasa~ said:
Not at all

Moofasa~ said:
of course not

Moofasa~ said:
Just because they temporarily used portions of the GoldSource engine to test certain portions of the Source engine doesn't mean the "Source engine had a development path forking directly from the Quake engine".

Perhaps I haven't made my position clear to you (although I'm still trying to figure out how this is even possible). But once more with feeling, no. Is that in simple enough terms for you to understand?
 
Yes thanks. I asked a yes/no question and that was the first time you provided a boolean response (as is required for a yes/no question) to that particular question. When you said "of course not" (which was the only other time you provided a boolean response) you directed me to replace a significant section of the question with different words, therebye not answering the question I had asked.

At this point it is clear to me that this is quickly becoming an argument of semantics if you do not believe that a fork of a fork of the Quake engine is the same as having a "development path forking directly from the Quake engine". This should be quite obvious if you consider that a fork is a development path, so I could just as well have said it was a "fork forking directly from the Quake engine" which is indisputable.

I have very little interest in arguing semantics over the internet so I'll leave it at that.
 
if you do not believe...

Because you haven't made that connection yet. How about instead of saying the same stupid line for four posts in a row, you answer my question:

Moofasa~ said:
How can the Source engine be "ultimately based" on the Quake engine if it doesn't share any common code?
 
I think source engine might still have code from quake 1. They bought the license for quake 1 and modified it for hl1 and i think theres still stuff from halflife 1 in the source engine but its not much but its there like engine code but not really graphical.
 
While I can see that there has been some heat arising from the forking of the forking of etc... Here's my thought on this.

Quake -> HL 1 = True, as seeing it's a fact for sure...
HL 1 -> HL 2 = True, however is it true that quake codes has been some how transferred from HL 1 to HL 2? I think this is more of a question in practice that needs to be answered before we can say that HL 2 is generally based partially on quake codes.

Now can we answer this? If so then fork of the forking is making a sense but however if not then it holds no water.

And I voted Source.
 
John Carmack said:
"I still take a degree of pride in the fact that there's still a bit of me, a bit of Quake I code, left in the Source engine. Try looking sometime and you'll probably still find comment fields from Robert Duffy and myself."

"I look through it and think, I wrote that 10 years ago. .. They had the Quake I license, built Half-Life on it and just kept building and creating on top of that. They bought out their license with us to have infinite distribution. There's a lineage and a character there and I'm real happy being a part of that."
Link. Anyone here have a source that'd be sufficient in contradicting this?
 
Link. Anyone here have a source that'd be sufficient in contradicting this?

Have you read any of the thread? :confused:

Read the book that comes with one of the special versions of Half-Life 2. Valve clearly states the Source engine is built from the ground up with no Quake code.
 
I've obviously read the thread having posted in it numerous times.

I find it strange that Valve book would say, quite specifically, that there is no original Quake code in Source (are you referring to "the making of" book?). That's not really something you'd say in passing. What does the passage say exactly?

I suppose there's no easy to way to confirm this either way. Either Valve's "lying", or John's "lying". Might as well just pick a side.
 
I find it strange that Valve book would say, quite specifically, that there is no original Quake code in Source (are you referring to "the making of" book?). That's not really something you'd say in passing. What does the passage say exactly?

I suppose there's no easy to way to confirm this either way. Either Valve's "lying", or John's "lying". Might as well just pick a side.

Well they didn't say verbatim "there's no Quake code", but they said during development they split Half-Life 2 into four (it might not be four as I can't remember exactly, but I think so) parts. Before they put all the parts together they tested each one with the GoldSource engine (to ensure each was functioning as expected). Eventually they brought all the parts together (one at a time) to form the final and complete version of the Source engine.

I don't think anyone is lying. It's not hard to conceive that John saw an early version of the Source engine (the leaked version) that included a couple of insignificant ID files, but were later removed in the final Source engine (as they were unneeded).
 
Graphics and physics wise I'd say Crysis is the best. That's not to say that Source pales in comparison. It's different, but it is a bit dated in the amount/quality of graphics that it can display next to Crysis. As for U3, I have only played Bioshock which had excellent graphics and was beautiful, at least on par with the newest source engine, but didn't have the level of realism or immersion of the Crysis engine at all. Things do seem more... fake or plastic in U3, though Bioshock still had awesome art direction and made it look good.

People that say the Source engine is based on Quake are talking about something they know nothing about. The Source Engine was developed using the knowledge they had from the extensions they did to the Quake engine. The Half-Life 1 engine had probably 20% in common with the Quake engine and was so much more advanced than the Quake engine with only a fraction of the original code that it could be called its own engine.

I'm going with the Source. It runs well and it looks realistic. I still think that only Crysis has the same realism that HL2 does, and it takes twice the horsepower to run it.
UE3 runs really well, but the textures look muddy and everything has a look and feel that looks like an odd drawing. It looks good, but there's no sense of "this looks REAL." None of the licensed games look realistic either, so I just think it's not capable of outputting the look and feel I want without more modification.
CryEngine 2 is probably the best overall, but right now it's a system hog that none of us can truly fully experience. Hold this poll again next fall and I bet you'll see it winning easily.
For now, the Source engine looks wonderful and will run perfectly on hardware from 3 years ago. The physics, textures, facial expressions, sound, controls, etc. are all perfectly tuned for current hardware and have been since its release. There's something to be said for getting a flawless running/looking game that everyone can play.
With Crysis, you're pretty much stuck using a DX10-level card unless you want everything to be a low-detailed mess or slideshow.

Weird. I ran the demo on my Opteron 170 + 7950GT card in XP32 and it ran quite smoothly with everything set to medium except textures and water set to high. :rolleyes: The game was still breathtaking and the best looking game I've ever seen and one of the most fun I've played in a long time.

Perhaps you were playing a different game?
 
I picked Source. I prefer the visual style and the amazing sense of movement in Source games – something that is lacking IMO from any of the UE3 games or Crysis and hearkens back to the feel of Quake. As far as I know UE3 is the only engine of the three where a licensed developer has brought a lawsuit against the parent for an alleged lack of support and tools leading to production delays. Cry2 is probably too new for this and Source has ended up being more of a Valve-centric platform (definitely not a bad thing) and they answer to no-one but themselves.

After playing through Portal with the developer commentary I came away with a whole new level of respect for what was achieved and the programming behind it. At least part of that should be attributed to how robust the engine and development environment are. Source is also absurdly scalable, making sure that the broadest possible range of customers can enjoy finished product, something that cannot be said about either UE3 or Cry2.

The new CoD engine is also pretty nice.
 
Source runs the best and still looks good, but Crysis is totally next gen.. I'd say Source since i can run it max and it still looks good, and that engine is almost 3 yrs old.
 
It will take a bit for hardware to catch up but cryengine 2 is definently the prettiest and will be for a while.
 
Best / Most Innovative / Market -Most adopted

Best: Prolly Unreal 3,
The best engine is one that delivers the most graphics and can maintain a high level of gameplay. Crysis may have more physics and detail but if the game is less than playable due to poor performance it can't qualify as the best engine.. *most people do not have the latest and greatest every week as new hardware is released*

Most Innovative: ID & Doom
First to the next gen level of design, Carmack always writes very good engine's, Dooms launch wasn't as good for many reasons.. the biggest being timeing.. They released pre physic's craze and pre release of gfx cards with real power *512 was just barely being though of to be released*

Market: Epic, Source > ID, > Cry2eng
Epic delayed unreal 3 when their engine wasn't completed in time to compete with Doom3 engine, this worked in their favor as the next gen of cards and trend hit.. and unreal3 was able to more fully integrate these features.. it is no secret that this has taken place. Also due to the poor release time and im sure other factors. the unreal 3 engine had better liscencing relations over its only compition ID.. Time will tell which company benifits more, for now its EPIC.

Cry2engine is not a factor.. like farcry before it.. engine adoption has been lacking for the company. They look great dont get me wrong, but making a game look good, and making it run well for the consumer base and being adopted by other game companies are too differant things.

Source (valve) : Started from the orginal quake engine, half -life.. Popularity from Half life+mod Counter-strike ultimatly proved the launching ground for the company. (nobody can deny that).. Source was the updated engine and through high recognition of the fanbase, and ease of use of the new content delivery Steam, has proved Source to be a large force with both delivery and excelent graphics. Making adoptation for other game companies easy as part of the STEAM network.


if i had to say something now.. Unreal 3, because its more advanced than source and will run good while maintaing the higher quality. along with the benifit of a differant time period of launch on the technological side.

but... i still think Carmack writes the best engine code in the industry. There are just people that are that good. and hes one of them. Unfortunatly the second guy needs only to observe/copy/adapt a differant way, and can definatly make a bigger impact :)
 
Call of duty 4 is based on Quake 3 engine. Places say it isn't but its like this Quake 3(cod1+uo)--->Call of duty 2---->Call of duty 4. You can still use quake 3 commands and all types of other stuff in the console still. It scales really good. Still a pretty good engine :)

As for counter strike, If it wasn't for a couple highschool kids moding halflife 1 we wouldn't even be seeing steam alive today practically ;)
 
Cry2engine is not a factor.. like farcry before it.. engine adoption has been lacking for the company. They look great dont get me wrong, but making a game look good, and making it run well for the consumer base and being adopted by other game companies are too differant things.

It seems you're not up to speed, but CryEngine 2 was already licensed, at least half a dozen times, before Crysis was even released. And that number will surely increase, when the full game is out.

darkar said:
Source (valve) : Started from the orginal quake engine, half -life.. Popularity from Half life+mod Counter-strike ultimatly proved the launching ground for the company. (nobody can deny that).. Source was the updated engine and through high recognition of the fanbase, and ease of use of the new content delivery Steam, has proved Source to be a large force with both delivery and excelent graphics. Making adoptation for other game companies easy as part of the STEAM network.

Source was only licensed for a couple of titles, outside of Valve, so it definitely is not a hit as you try to make it out, in terms of adoption by other companies.
 
I'm going with the Source. It runs well and it looks realistic. I still think that only Crysis has the same realism that HL2 does, and it takes twice the horsepower to run it.
UE3 runs really well, but the textures look muddy and everything has a look and feel that looks like an odd drawing. It looks good, but there's no sense of "this looks REAL." None of the licensed games look realistic either, so I just think it's not capable of outputting the look and feel I want without more modification.
CryEngine 2 is probably the best overall, but right now it's a system hog that none of us can truly fully experience. Hold this poll again next fall and I bet you'll see it winning easily.
For now, the Source engine looks wonderful and will run perfectly on hardware from 3 years ago. The physics, textures, facial expressions, sound, controls, etc. are all perfectly tuned for current hardware and have been since its release. There's something to be said for getting a flawless running/looking game that everyone can play.
With Crysis, you're pretty much stuck using a DX10-level card unless you want everything to be a low-detailed mess or slideshow.

Crysis Was made to be ran on DX10 Graphics cards,new games are going to be made for new hardware.I say That The CryEngine 2 is the best overall.
 
Sorry but Valves marketing is SHITE.

They cannot get a release date accurate to within about a year of the actual release date. They totaly arsed up the whole back box/orange box thing where they basically made loads of gamers buy HL2 and Ep1 again.

I really can't think of any worse marketing of recent history.
 
Sorry but Valves marketing is SHITE.

They cannot get a release date accurate to within about a year of the actual release date. They totaly arsed up the whole back box/orange box thing where they basically made loads of gamers buy HL2 and Ep1 again.

I really can't think of any worse marketing of recent history.

I thought it was greate marketing, especially since the orange box was released on the Xbox360. Thought it was the best gaming package money could buy especially since you could get it for $39.99 on the PC at select stores.
 
Crytek really shot themselves in the foot with Cry Engine's architecture support. It's fantastic for outdoor stuff, but it was like they didn't really give much though to buildings and indoor environments. If you want to market an engine, you need to make it useful for things beyond the one game you want to make with it. I'm pretty sure that's why Cry Engine 1 didn't get licensed.
 
Crytek really shot themselves in the foot with Cry Engine's architecture support. It's fantastic for outdoor stuff, but it was like they didn't really give much though to buildings and indoor environments. If you want to market an engine, you need to make it useful for things beyond the one game you want to make with it. I'm pretty sure that's why Cry Engine 1 didn't get licensed.


Are you referring to the first or second CryEngine?
 
Ok, here's a pano comparison of modded cfg vs "high" (stock) in Crysis

Modded CFG (DX9, HIGH)


No CFG mod (DX9 HIGH)


Notice the foliage has edge AA and a better look, water also has a more natural look to it, it also has CA. The background forest also has more hazy look, it looks more like DX10 on Veryhigh

And some other screens using custom cfg:
http://robbiekhan.co.uk/root/games/crysis/configmod/splode.jpg
http://robbiekhan.co.uk/root/games/crysis/configmod/jump.jpg

http://robbiekhan.co.uk/root/games/crysis/configmod/gravity.jpg
^ Here I threw a guy into the water and followed him in, I was surprised that Crytek had implemented real world underwater physics too! the guy floated for a bit then slowly started sinking (torso first) and his arms/legs remained higher than the rest of his body until he hit the floor, thumbs up!

The more I play this the more fun I have in SP mode, the beach is just excellent to mess around in and it's such a beautiful level.

As for CryEngine2 vs Unreal Engine 3? well this is all I have to say to that:


Source engine just doesn't fit in between these 2. Sure the physics are excellent but it doesn't touch Crysis or the weather and atmosphere of UE3.
 
Back
Top