John Carmarck bothered with Next gen MProcessor Consoles

london-boy said:
Jov said:
It might break open new techniques or expand on some existing but under utilized methods.

:| Such as...?

He hasn't magically "found new techniques" on PC, what makes you think he can do that on consoles?

London-boyie... Tell me, how do developers (be it artist, programmer, etc…) come up with new techniques/ideas/methods for creating games?

In this industry creativity is a key to success. Though success is not guaranteed, but does increase your chances significantly.

Anyone working/developing on a particular platform will improve their skills on the said platform. Who's to say Carmarck (or anyone) working on consoles can't come up with something new?

As for your "magically..." comment, it is your interpretation not necessary what stated.

If JC does change his primary platform of choice to consoles, it might be very interesting.


Anyone working/developing on a particular platform will improve their skills on the said platform. Who's to say Carmarck (or anyone) working on consoles can't come up with something new (or different)?

As for your "magically..." comment it is your interpretation, not necessary what stated.
 
Do u mean "new things" technically or gameplay-wise then? "New things" is very general as a term, and PERSONALLY i find it hard to believe Mr Carmak will find anything new gameplay-wise. As for technology, all he can do is use pretty techniques which are not exactly "new", and focus on them. Look at Doom3, nothing in the game is "new", the opposite in fact. The way the effects are implemented is impressive though (from what i've seen).
 
london-boy said:
Look at Doom3, nothing in the game is "new", the opposite in fact. The way the effects are implemented is impressive though (from what i've seen).
AFAIK Carmack developed some novel technique for DOOM3 technology,
such as the algorithm now widely recognized as 'carmack's reverse'.

ciao,
Marco
 
Jov said:
london-boy said:
Jov said:
It might break open new techniques or expand on some existing but under utilized methods.

:| Such as...?

He hasn't magically "found new techniques" on PC, what makes you think he can do that on consoles?

London-boyie... Tell me, how do developers (be it artist, programmer, etc…) come up with new techniques/ideas/methods for creating games?

In this industry creativity is a key to success. Though success is not guaranteed, but does increase your chances significantly.

Anyone working/developing on a particular platform will improve their skills on the said platform. Who's to say Carmarck (or anyone) working on consoles can't come up with something new?

As for your "magically..." comment, it is your interpretation not necessary what stated.

If JC does change his primary platform of choice to consoles, it might be very interesting.


Anyone working/developing on a particular platform will improve their skills on the said platform. Who's to say Carmarck (or anyone) working on consoles can't come up with something new (or different)?

As for your "magically..." comment it is your interpretation, not necessary what stated.
I think it's Carmack, not Carmarck.

Dooms and Quakes, a couple of capable PC fps engines and gameplaywise ok'ish shooters and he's considered a god.

I think he's got name more because PC games need(ed) an idol, a face. Like consoles have their Miyamotos, Suzukis, Yamauchis, Kojimas etc...
PC games had Romero, that Shiny guy... now there's only more or less just Carmack and that Unreal guy.
As the teams producing games get bigger and bigger, there will be less of these celebrity programmes, instead the 'faces' in game business will be the 'director' types, such as Hideo Kojima, Kazunori Yamauchi, Yu Suzuki...
 
london-boy said:
Do u mean "new things" technically or gameplay-wise then? "New things" is very general as a term, and PERSONALLY i find it hard to believe Mr Carmak will find anything new gameplay-wise. As for technology, all he can do is use pretty techniques which are not exactly "new", and focus on them. Look at Doom3, nothing in the game is "new", the opposite in fact. The way the effects are implemented is impressive though (from what i've seen).
hun, its Carmack.
 
nAo said:
london-boy said:
Look at Doom3, nothing in the game is "new", the opposite in fact. The way the effects are implemented is impressive though (from what i've seen).
AFAIK Carmack developed some novel technique for DOOM3 technology,
such as the algorithm now widely recognized as 'carmack's reverse'.

ciao,
Marco

Isn't there a presentation from 3dlabs that outlines the core of depth-fail stencil shadowing a year or two before Carmack?
 
nAo said:
AFAIK Carmack developed some novel technique for DOOM3 technology,
such as the algorithm now widely recognized as 'carmack's reverse'.

ciao,
Marco

The actual theory behind "Carmack's reverse" is as old as the hills. The description I've read of him "discovering" the technique is all very interesting but mostly just a case of him going around in circles for a while until he proved the basic mathematical fact that when tracing a line from some point to infinity, it doesn't matter what direction you go in, you'll get the same number of intersections between the line and any closed volumes within the space. If anything the hoops he jumped through to come up with it actually just make it look like (to me at least) he didn't particularly understand the problem to start with.

It's neither new, nor particularly novel.

I take my hat off to the guy for his acheivments and I'm sure he's come up with many original solutions to various problems he must have encountered - but some of the stuff people attribute to him is just plain rediculous.

"He's not the messiah - he's just a very naughty boy"
 
I googled around with Carmack's reverse and I found that probably JC wasn't the first one to came up with the main idea behind it (in fact it seems Hoppe patented it in 1999) but it seems also that before JC no one knew how to manage shadow volumes intersecting the view frustum in the general case. This is novel to me :)

ciao,
Marco
 
nAo said:
I googled around with Carmack's reverse and I found that probably JC wasn't the first one to came up with the main idea behind it (in fact it seems Hoppe patented it in 1999) but it seems also that before JC no one knew how to manage shadow volumes intersecting the view frustum in the general case. This is novel to me :)

ciao,
Marco

Well the idea behind it extends back much further than that, though it doesn't surprise me in the least to find out that someone managed to patent a variation on it (not having read the patent in question I'm not clear exactly was aspect of "counting" hoppe managed to patent).

IIRC there was a similar trick with the Amiga blitter for doing filled polygons. Instead of using the filling mode to do a horizontal fill, you could instead perform the same operation vertically by having a blit XOR the previous line into the current one. This exploits the same basic property as the volumetric equivalent, i.e. changing the direction of polygon scan-conversion does not change the result.

Can't remember exactly why that was useful at the time though - perhaps it was just more efficient to draw the edge pixels in that mode due to the plane-oriented display... fairly sure the blitter DMA usage / speed wouldn't be affected...

OK, veering off-topic into nostalga :)
 
akira888 said:
nAo said:
london-boy said:
Look at Doom3, nothing in the game is "new", the opposite in fact. The way the effects are implemented is impressive though (from what i've seen).
AFAIK Carmack developed some novel technique for DOOM3 technology,
such as the algorithm now widely recognized as 'carmack's reverse'.

ciao,
Marco

Isn't there a presentation from 3dlabs that outlines the core of depth-fail stencil shadowing a year or two before Carmack?

I believe S3 presented, and then patented it (in the US), prior to Carmack. This leads to the interesting situation that you can use the technique freely anywhere except in the US.

[FWIW the US is the only place I know where you can publicly present something and then patent it up to 1 year later].
 
nAo said:
I googled around with Carmack's reverse and I found that probably JC wasn't the first one to came up with the main idea behind it (in fact it seems Hoppe patented it in 1999) but it seems also that before JC no one knew how to manage shadow volumes intersecting the view frustum in the general case. This is novel to me :)
ciao,
Marco

That must be a patent for something else. FWIW the general idea of using the stencil buffer to implement Crow's shadow volumes has also been patented - by SGI I believe.
 
rabidrabbit said:
I think he's got name more because PC games need(ed) an idol, a face. Like consoles have their Miyamotos, Suzukis, Yamauchis, Kojimas etc...
PC games had Romero, that Shiny guy... now there's only more or less just Carmack and that Unreal guy.

To some extent this is probably true.

However, John Carmack has been instrumental in bringing several mile-stone titles to gaming. Whether they were "good" or not is beside the point - they were seminal/significant.

Another major reason he is held in such high regard is that he has been very open and sharing with his knowledge. Not only with other programmers and people in the industry, but also with (semi-creating you could say) the mod community. Furthermore, id has been very generous with their intellectual property, making source code freely available, dropping CD-checks making life easier for their end users et cetera. He has also been very supportive of both the MacOSX and Linux, despite marginal economic justification.
He has simply earned a lot of respect, in a lot of places.

Add that he very carefully tries to be correct when making public statements, and clarifies/corrects himself publicly if for instance he finds that he has been too ambigous, and you have a man that people listens to when he speaks.

He may have become an icon, but there are good reasons for it.
And, IMHO, he is one of the better icons around.
 
Back
Top