Sorry your browser is not supported!

sometimes i have to wonder if you were kicked in the head by a horse when you were younger.

you know that you can load a DirectX 8.1 Model format in OpenGL using the DirectX 8.0 library ... now these models can be setup with animations and with identical features.
but the one you load in OpenGL with the 8.0 library won't be able to use BlendWeights/VertexColours/PointSizes/MultipleUV Maps/etc...

as i said i'm not going to aruge what you want to anymore so i'm just going to see if i can open your eyes just a faction so you can see the truth staring you directly in the face.

"sometimes i have to wonder if you were kicked in the head by a horse when you were younger."

I'd say the same of you.

"you know that you can load a DirectX 8.1 Model format in OpenGL using the DirectX 8.0 library ... now these models can be setup with animations and with identical features.
but the one you load in OpenGL with the 8.0 library won't be able to use BlendWeights/VertexColours/PointSizes/MultipleUV Maps/etc..."

Your analogy is incorrect and inconsistent with what you have stated earlier. What your saying is that, basically, one shader is a subset of the other. One shader has many features, but the other shader has those features as well as more. That is why the former shader can be compiled as the later and the later can not be compiled as the former. Since Cg compiled as HLSL then HLSL must be the shader format that has all of Cg's features and more. That extra "more" being what alledgedly prevents it from compiling as Cg.

But that hasn't been what you were arguing was it? What you were arguing was that it was Cg that was the superset with more features than HLSL with statements like:


If what you said previously was true then it would be Cg that couldn't be compiled as HLSL and HLSL compiling just fine under Cg. The fact of the matter is that both compile just fine as one or the other. You just don't want to admit you are wrong.

"as i said i'm not going to argue what you want to anymore "

Won't argue or can't? By the way, you never said you weren't going to argue anymore in any of your previous posts.

"so i'm just going to see if i can open your eyes just a faction so you can see the truth staring you directly in the face."

If making half baked analogies and telling inconsistent falsehoods is your way of making me see the "truth" then you are putting up a mighty big effort.

I take that by now you see that I'm right and there is no way to argue your position given the amount of evidence I've put forward. This is just your way of backing out while trying to look like you were in the "right" all of the time. Whatever. I don't really care anymore. Though I'm am still curious as to what HLSL shader didn't compile under the Cg compiler so let me ask one more time. Why don't you post the code that you used? You are confident that it won't compile aren't you?

no just no point in arguing with a brick wall ... at the end of the day i've never seen a single working shader example from you and any real proof other than you have a good ability to quote text and take examples from said texts to backup claims.

the same could be said about the bible, 20 different people could extract a section of text and get 20 different answers ... which one is right about it is still an arugment fought today.

no point in continuing to argue about the symantics of languages with someone who won't listen, its like trying to convert a Jew to a Christian - you'll believe what you want, and i'll believe what i want so give it a rest.

In other words Raven, you lost the argument, and you can't take it. LMAO

doesn't look like i've lost, i'm just tired of arguing for once ... infact i'm completely sick of trying to stress a point to you guys, i'll let you believe whatever the hell you want - and when something doesn't work the way you think it does just don't come crying to me wondering what the hell is wrong cause you ain't gonna get any sympathy.

if a select few of you feel that it is your points to always stand up and try to make out i'm wrong - then fine think i'm wrong. Isn't going to stop the facts from being right.

The real problem with what you were saying Raven is that Neophyte backed up his argument with examples / links etc. In your case you provided none of these. How on earth I am supposed to believe you if you cannot produce any evidence to support your side.

"doesn't look like i've lost"

Well, for the reasons I stated above, yes it does.

I am not an expert on the subject, and I don't pretend to be, but as a newbie looking at this, I would come away knowing Neophytes view of this as the truth.

thats funny because i've written and entire Cg and Asm shaders in other threads, and they were just copied from text which i barely understand - they were made from scratch.

if you want to believe Neo then just reread the reply you got from nVidia, that is an ambigious answer - common PR practise is to give ambigious answers when you don't want to countermand something which has already been stated otherwise or to just agree with the user/emailer (ie you).

although it clearly says that things are similar, at what point does it say they are the same as Neo would have you believe? at what point did she say that the structures are also identical as Neo again would have you believe?

neo can quote all he likes from manuals, it doesn't mean that he understands the subject at all ... it just means he can quote to enhance his case. Thats not my style to do and nor will it ever be, i have released working shaders into this community which aren't just there but are actually 100% self developed NOT something i've taken from a website, or from an SDK. I've explained to alot of users how the shaders work, and what they're about - especially Bullseye who is actually begining to understand shaders alot better now and howto use them effectively as well as actually use them in the first place.

i don't need a big fat manual to quote from to explain to people why you would create a Per-Pixel-Lighting effect within a Vertex Shader or why you would use the Vertex Shader and not a Pixel Shader.

this is just a subject i'm stubling into getting all of the tutorial books and crap like Neo has been, i am yet to see any solid evidance from Neo to actually contradict what i've said ... he's even posted that the example i told him DID NOT compile first time, but he seems to think they worked idencially.

i wouldn't mind knowing was he using C/C++ with OpenGL or DirectX, which interface was he using Dx8, Dx9, Core or OpenGL?
can he show screenshots of his results and the compiler log to back up his claims?

at the end of the day a book is just a book written by people who believe they know enough, there are so many errors in the Cg help 1.0 that it was COMPLETELY rewritten for 1.1 and is being rewritten again for 1.2 ... at the end of the day the coders are not writers as the writers are not coders, and misconceptions will happen between the two. If you don't believe that just checkout the developers forums for Cg and notice how many people complain that thier code won't work 100% of the time just to find out that something the manual said or even worse did not mention was the actual fault.

i'm sure if you went through the DBP manual you'd find the exact same deal going on, some of the functions are either poorly commented or sometimes innaccurate to how they work. Not much, but just enough to throw off a newbie on how it should work and then they wonder why thier code doesn't and come on here.

the fact that you'd see Neo's POV as the truth and not to question because he has it from source is the exact reason i was actually arguing in the first place ... but now i don't see the point, once people make the mistakes then they'll see what i've been saying was right - and there isn't anyway of getting around it.
You can fool people but not a Compiler

@Raven

"no just no point in arguing with a brick wall ... "

...or someone who is right.

"at the end of the day i've never seen a single working shader example from you "

What? What about the refraction shader I posted on the previous page? Or did that just magically appear out of nowhere? Come to think of it I haven't seen a single working shader out of YOU Raven.

"and any real proof other than you have a good ability to quote text and take examples from said texts to backup claims."

And I haven't seen any real proof that you know what you are talking about even in the slightest. I don't what things are like where you come from but from where I come from when you make an outrageous claim you back it up with proof, not BS and hot air.

"the same could be said about the bible, 20 different people could extract a section of text and get 20 different answers ... which one is right about it is still an arugment fought today."

But, unlike the bible, Cg is barely a year old and it's creators our around today to answer our questions and explain what they mean. Cg is a programming language not a religous text. It is not open for interpretation. It either compiles or it doesn't.

"no point in continuing to argue about the symantics of languages with someone who won't listen,"

You know I wonder if you ever truly grasp the irony of it all when you say things like this. I know I do.

"its like trying to convert a Jew to a Christian - you'll believe what you want, and i'll believe what i want so give it a rest."

In other words, you don't care about the facts if they don't conform to your world view. Since I know too much about this you can't BS me about it like the others so you run away pretending I'm the ignorant one all the while claiming that you were right all along. Sorry, but I take a lot of offense at that.

@Rob K

"The real problem with what you were saying Raven is that Neophyte backed up his argument with examples / links etc. In your case you provided none of these. How on earth I am supposed to believe you if you cannot produce any evidence to support your side."

This is what I've been telling Raven all along and I'm glad I'm not the only one who sees this. It makes you wonder if he really believes what he says or is he just screwing around trying to annoy me.

"I am not an expert on the subject, and I don't pretend to be, but as a newbie looking at this, I would come away knowing Neophytes view of this as the truth."

Thanks for seeing my POV. I was starting to wonder if I was getting my point across effectively what with Raven's repeated denial of the facts and him constently ignoring the validity of my examples.

If you need any more help with shaders then just ask. I'd be glad to help.

@Raven

" doesn't look like i've lost, i'm just tired of arguing for once "

Or you realized that you were wrong and can no longer defend your position.

"... infact i'm completely sick of trying to stress a point to you guys, i'll let you believe whatever the hell you want "

That's a pity. I was hoping you would post another one of your patented outrageous claims. Guess, I'll have to get my amusement elsewhere.

"and when something doesn't work the way you think it does just don't come crying to me wondering what the hell is wrong cause you ain't gonna get any sympathy."

Why would any one come crying to you? You haven't a friggin idea what is going on so how would you help?

"if a select few of you feel that it is your points to always stand up and try to make out i'm wrong"

You'll be shocked to know that I'm actually not one of them. I only argue with you when you argue with me. I don't actively seek you out. Check both of the threads we had arguments in. As far as I can tell you fired the first shot so to speak. I know I've held my tounge on more then one occasion when you post something goofy or inaccuarte so this isn't a case of me picking on you whenever you post.

" then fine think i'm wrong."

I don't think you're wrong. I know you're wrong.

"Isn't going to stop the facts from being right."

True. But in this case the facts aren't on your side. Unless of course you would like to offer up a few links or snippets that prove otherwise.

@Raven

"thats funny because i've written and entire Cg and Asm shaders in other threads,"

Really? Which ones? This ought to be entertaining.

"and they were just copied from text which i barely understand "

I love it when you're honest.

" they were made from scratch."

By someone else.

"if you want to believe Neo then just reread the reply you got from nVidia, that is an ambigious answer "

What part of "NVidia's Cg language and Microsoft's HLSL language are very close in terms of syntax. " is ambigious? You are in serious denial Raven.

"common PR practise is to give ambigious answers when you don't want to countermand something which has already been stated otherwise or to just agree with the user/emailer (ie you)."

No, it isn't. Anyway, how would they know what position Rob K held in order to agree with him? From the contents of the message it appears that Rob K just asked them how similar they were. There is no indication that they knew what position, if any, he held on the matter. Even if they did, why would they send that to him just to "agree" with him? What would they possibly have to gain from lying to him? They would, however, have quite a bit to lose if they lied to him and he found out(and with a lie that big it would be pretty easy to found out). It doesn't pay to have a dissatisfied customer roaming around telling people that they have no idea what they are talking about.

"although it clearly says that things are similar, at what point does it say they are the same as Neo would have you believe?"

I never said they were completely 100% identical. In fact, I always prefaced every comparison I made between them with "basically" or "almost" identical. I even explained what was different about them(the inclusion of the fixed data type in Cg). Cg and HLSL are basically identical. What will compile in one will almost always compile in the other except for a few rare cases. This has always been my position on the matter.

"at what point did she say that the structures are also identical as Neo again would have you believe?"

When she(or he) said that they were very similar. Still, this is a far cry from all of your claims that they were a lot different. If they are a lot different, Raven, then explain why the people at the Nvidia developer relations say that they are very similar.

"neo can quote all he likes from manuals, it doesn't mean that he understands the subject at all ... "

But if I didn't understand the subject then how would I know that you were wrong in the first place?

"it just means he can quote to enhance his case."

Or that I might actually know what I'm talking about.

"Thats not my style to do and nor will it ever be,"

You mean back up what you say with evidence and/or intellegent insight? Well color me suprised.

"i have released working shaders into this community which aren't just there but are actually 100% self developed NOT something i've taken from a website, or from an SDK."

And where are these phantom shaders of yours? Inquiring minds want to know.

"I've explained to alot of users how the shaders work, and what they're about "

So have I. But the difference between you and I is that my information is accurate while yours is not as evidenced by this thread and the other one.

"i don't need a big fat manual to quote from to explain to people why you would create a Per-Pixel-Lighting effect within a Vertex Shader or why you would use the Vertex Shader and not a Pixel Shader."

Actually you would use both, but that is another thread topic.

"this is just a subject i'm stubling into getting all of the tutorial books and crap like Neo has been,"

I noticed.

"i am yet to see any solid evidance from Neo to actually contradict what i've said "

Haven't seen or refuse not to see? And don't you think it is a bit hypocritical to demand evidence when you don't forward any at all.

" he's even posted that the example i told him DID NOT compile first time, but he seems to think they worked idencially."

I never posted that. Go back and reread my post. I said:


"i wouldn't mind knowing was he using C/C++ with OpenGL or DirectX, which interface was he using Dx8, Dx9, Core or OpenGL?"

And once again you demonstrate your ignorance of what the runtimes really do. Though it is nice that you finally recognize that Cg doesn't bypass the 3D APIs you forget that the Core runtime must be included with another 3D api runtime in order for the Cg programs that it compiles to be used. So it is not a matter of "Dx8, Dx9, Core or OpenGL" but the Core and one of the 3D apis. You see this is why I have a hard time believing that you ever really programmed with these(at least with C programs for Nvidia as you claimed anyways). If you had really written a program that used the Cg "interface" then you would have known this. For it would become immediately obvious that something was wrong when none of your shaders were displayed!

"can he show screenshots of his results and the compiler log to back up his claims?"

I don't have a website nor do I have the inclination to learn HTML and get one just to post some screen shots when all you will do is deny that they are HLSL and say that they are both Cg. Or I faked them or whatever. Hell, you haven't even posted what shader you used that didn't compile! If you can't cut and paste a few lines of text then why should I go to all of that effort just so you will say that I am still wrong and my screenshots don't prove a thing? Your hypocrisy, Raven, is stunning.

"at the end of the day a book is just a book written by people who believe they know enough,"

Are you saying that the people at Nvidia don't know what they are talking about? You are one heck of a mind job Raven I'll grant you that. How you can wrap your head around this distoration is beyond me.

"there are so many errors in the Cg help 1.0 that it was COMPLETELY rewritten for 1.1 and is being rewritten again for 1.2"

I'd ask for proof that this were so but you just ignore me when I ask that question. Anyway, the language has changed a bit since 1.0 so yes there would be some modifications but that doesn't necessarily mean that they were wrong at the time just that the language has changed and that they are no longer accurate. Though I won't deny there were a quite a few errors the exsistance of those errors doesn't mean that everything in the manuel is wrong or, more specifically, that what I said was necessarially wrong.

I know I'll just be wasting my time but I will once again ask you to provide a example(specifically the shader that you had that didn't compile "first time".) of a major incompatiablity between HLSL and Cg and how that makes them "ALOT different".

"at the end of the day the coders are not writers as the writers are not coders, and misconceptions will happen between the two."

I take it you mean "miscommunication", but I believe that that is what proof readers are for. Correcting discrepencies that result from it no? One would think that a company as large as Nvidia would at least have one on staff wouldn't you think?

"If you don't believe that just checkout the developers forums for Cg and notice how many people complain that thier code won't work 100% of the time just to find out that something the manual said or even worse did not mention was the actual fault."

True, but that still doesn't prove that what I said was wrong(Cg and HLSL are basically identical) and that what you said was right(Cg and HLSL are ALOT different). There may be errors in the language documentation but are they so major that something like a simularity in syntax which appears to occur between two languages isn't really there? If that is true then how would anyone get any shaders running at all?

"i'm sure if you went through the DBP manual you'd find the exact same deal going on, some of the functions are either poorly commented or sometimes innaccurate to how they work. Not much, but just enough to throw off a newbie on how it should work and then they wonder why thier code doesn't and come on here."

True, but then again Darkbasic software isn't a major international corporation with thousands of people on staff. They don't have the money to hire professional writers to do their documentation like Nvidia does.

"the fact that you'd see Neo's POV as the truth and not to question because he has it from source is the exact reason i was actually arguing in the first place "

Huh? That made no sense whatsoever. Bad grammer aside, let me see if I can figuare out what you just said. The first part of "the fact that you'd see Neo's POV as the truth" seems pretty clear, but the second part of "and not to question because he has it from source" is pretty garbled. I take it that the "source" is the Cg language specificaton or runtime manual. It appears that what you said was that you were arguing with me because you knew that Rob K wouldn't question my sources when I quoted them so it was up to you to somehow correct me otherwise Rob would be misled. Now that Rob has sided with me on this you don't see any other point in arguing anymore because you have failed to win Rob over to your side correct?

"... but now i don't see the point, once people make the mistakes then they'll see what i've been saying was right- and there isn't anyway of getting around it."

Deny it all you want Raven but I'm right on this one. Even you know that.

"You can fool people but not a Compiler "

So where is this HLSL code you keep telling me about? Oh, thats right, I forgot that backing up what you say just isn't your "style."

oh look the master of quoting is back... choping up my post out of context as a tabloid would to put his point across.

that Shader you posted, would you like me to tell everyone where it came from or would you like to tell everyone that you copied it word for freaking word?
would you like to also tell everyone why you couldn't convert it to asm for everyone to use even though the Cg Compiler can achieve it, and oh whats this so can RenderMonkey ... Hmm go figure, but wait a minute don't you know Cg better than me to tell me about how it works and is structures and i mean you've been using RenderMonkey for alot longer than me so how come you wouldn't know it could do that too?

oh yeah i forgot, its because your blagging like you know everything but you know jack. Nothing within this arugment is your own, all you've done is taken from helpfiles and such throughout the web little excerts to get your point shown but doing what a tabloid would ... i mean christ if anyone one wants a better example your doing this all they have to do is read the post you've taken your last bunch of quotes from and see that how you've spliced up what i've said was taken to context your arugment and not what was really said.

unlike you i'm not sitting here trying to bullshit my way to win an arugment, nor will i ever ... this is why i rarely if ever use quotes to further my point - because it is upto people to read the information for themselves and come across what is said. I could quote your 12ways from Sunday and come up with valid arguments against you as well, but i don't need to becaus although you can blag what you know and fool people as i said you can't fool the compiler (that is if you even use the compiler)

the way i see it here all you want to do is continue to argue your point, even once i've said "fine believe what you want" which as Rob mentioned would really show you've won the argument you still wanted to push for further conflict.

if you really believed what you were saying then you wouldn't have posted again its as simple as that.

Look, I'm incrediably tired at the moment. I've had a busy day. I'll get to the rest of your post tommorrow. For now i'll just leave you with this that I stumbled across earlier today.

http://www.cgshaders.org/articles/interview_nvidia-jul2002.php

Just to let you all know that apart from the arguments here. DBPro will be Direct X 9 compatible for Patch 5. This is not a rumor but is fact.

The conversion was started yesterday on direct X 9 (Yesterday being Tuesday, 5/27/2003). Lee is hoping to be done with converting all the DLL's that need converting by today.

After this he will work on redoing the shading with .FX files for Direct X 9 shading...

The advantage of FX shaders being that you can put a lot of the info into the shader rather than making the end user enter lots of confusing data via SET VERTEX SHADER CONSTANT etc.

keywords meaning that they are not right now like you claimed... i'd start reading your own quotes.

and thats great news Blitz2pro i'll be looking forward to what i can do to edit up my Cg DLL as i'm hoping to release that alongside patch5

Hi guys. I'm still kinda lost with the cg thingy. Does it mean something like:

I code an "effect" using one of the tools mentioned earlier, and then later on "Load" the effect and use it in dbp?

Now, is that right? Or is that just a misconception? 10x!

Cg is a much easier way of creating shaders.

Shaders are completely customisable special effects such as bump mapping or realisitc water (reflection + refraction)

You write them using Cg (a programming language) and then load them in DBP and apply them to an object. Currently it is via CREATE VERTEX SHADER FROM FILE and CREATE PIXEL SHADER FROM FILE.

Currently DBP requires ASM shaders. This means confusing machine code.

Cg is much easier and the effect can be used more easily in DBPro because you don't have to set lots of confusing variables in order to get the shader to work, the shader can incorporate them automatically.

Rob you still have to declare all the data you use just like a standard asm shader ... i don't know where you get the misconception that you don't.

It is just a language which makes everything more sane, as to a user it makes more sense to see.
All the real advantage is, is being able to do the entire shader itself in the code rather than doing just the asm effect in runtime and then coding the actual program for the shader elsewhere.

"Rob you still have to declare all the data you use just like a standard asm shader ... i don't know where you get the misconception that you don't."

I think I misread / misinterpreted something Lee wrote in his diary.

probably ... i mean you still have to declare all the data you use throughout as you can't load it within the Cg/Fx programs.
though it is admitidly a simpler way to setup things, if he's done it right could could probably just use declared values.

so effectively rather than compiling the Cg you'd just pass the values through because its compiled into the main program, but if you use the compiled code or .fx you still have to set shader values

so depends if it is runtime stuff or compiled stuff

OK, but maybe DBP could be setup so that if the Cg constants / variables were given certain specified names or numbers or something then it would automatically assign the necessary vectors / matrices to them. This is just an idea though.

yeah that would be cool
if Lee's actually programmed the HLSL directly into dbp then we might see actually using them just like you can in DirectX within the actual code itself.

you'd just start like



or something like that

This rather pointless thread contains:

3162 Lines
21,366 Words
107,887 Charecters

If printed it would require 70 pages of A4 paper.

One cannot help thinking that the sheer effort expended in such a manner could have been better channelled into some coding project.

Darkheart

Whoa! - That is nothing like what I had in mind, I think you misinterpreted my idea.

No currently you might have to do this:



Several times to set everything up. It would be nice if that could be avoided.

@darkheart

In other words, you don't understand anything in this thread so you dismiss it as pointless. Isn't that a rather naive thing to do?

Some of us can multi-task as well, though that might explain the bugs and [sigh! dammit! >> back to the source code]

And besides, it would require 46 sheets of A4 paper, including graphics etc.

I understand the discussion I just don't see that it's getting anyone anywhere, two people have two seperate intractable positions, what more is there to discuss?

I think you must have not included all 77 messages or reduced the fonts to a very small size,either way does it really matter?

Darkheart

personally i code things whilst having an arugment about it... that way all the information i need is generally fresh in my mine.

i've just finished a month long development of and engine which uses nothing but nv30 (updating to nv35 on thursday when i get the new compiler to test hehee) formats

but yeah i know what you mean, stuff like the veiw matrix just seem like a waste of time declaring several times

you also missed of the first line being "global VEIW_MATRIX as dword"

but then everything is just enlongated really, however i do see your point about the veiw matrix and such which are common controls and no you shouldn't have to do any of that in Cg or Fx ... infact i like Cg because all you have to write is

#include <d3dtypes.h>
matrix4 matVeiw = D3DVIEWMATRIX ;

and there you have, same with projection, world, LH, RH, etc... and HLSL is pretty similar too.
i've just started planning everything for a Basic version of Cg, so probably see a simple alpha in a week (hopefully) ready intime for patch5 i'm really hopeing - that way it could be added.

You don't need to declare VIEW_MATRIX as a DWORD, you could just use 1 as an index value but I prefer proper variable names for clarity.

Been poring through Lee's diaries from the February to the current date, he was very focussed on making shaders very easy for the end user. Ie. One command to load a shader, one command to apply it, no need to set constants etc. Whether or not this will come to fruition is another matter.

well i think he'll need to actually be working towards a balance like with the data types.

have the simple stuff with one command functions and such, but when you need more complex things the option is there to expand to a more complicated area
if you over simplify shaders unforuntately they loose alot of thier major features.

@Raven

"oh look the master of quoting is back... "

The master of quoting...I like that.

"choping up my post out of context as a tabloid would to put his point across."

Other than with that amusing typo you made at the top of a previous post you wrote where exactly did I take your arguments out of context to get my point across? At what point did I misrepresent you?

"that Shader you posted, would you like me to tell everyone where it came from or would you like to tell everyone that you copied it word for freaking word?"

I already did you bloody idiot. Don't you know how to read?


I even underlined it in my original post. Your attempts at making me look bad are really getting rather childish.

"would you like to also tell everyone why you couldn't convert it to asm for everyone to use"

Ahhh yes, the refraction code. I had almost forgot. I was hoping to go through and flex my asm shader muscle and write it out by hand but I got caught up in work and arguing with you. I suppose that if I don't post it you'll rag on me so I guess I'll "cheat" and use the Cg compiler to save time. ::sigh:: I was looking forward to a challenging exercise.
Vertex shader.


Pixel Shader


"even though the Cg Compiler can achieve it, and oh whats this so can RenderMonkey"

Yes the Cg compiler can achieve this, but RenderMonkey can't. You can develop asm shaders in RenderMonkey, but sadly you can't compile them into a file. Shaders are stored in seperate XML files that need parsing. I consider it the One major flaw of the RenderMonkey design; you are unable to use the shaders you create in your own programs without a XML parser.

"Hmm go figure, but wait a minute don't you know Cg better than me to tell me about how it works and is structures "

That appears to be the case doesn't it?

"and i mean you've been using RenderMonkey for alot longer than me"

This is quite correct yes. I can't see how anyone would be under the impression otherwise what with your poor understanding of the profiles that it supports(see earlier in this thread) and the options that it supports. You did even admit earlier on that you really hadn't looked at it much remember?

" so how come you wouldn't know it could do that too?"

I think the real question is how could you "know" that it can? Have you ever converted any shaders in Cg into assembly with it? Somehow I'm pretty confident the answer to that starts with a N and ends with an O.

"oh yeah i forgot, its because your blagging like you know everything but you know jack."

No, I'm sorry but I don't know any jacks.

"Nothing within this arugment is your own,"

Really? Not even my own words? By that standard you too have nothing in this argument that is your own.

"all you've done is taken from helpfiles and such throughout the web little excerts to get your point shown"

And all you've done is throw a temper tantrum every time I showed you how utterly wrong you are. You haven't even posted a single example to back up your claims or contradict mine. You can't even take the 4 seconds it takes to cut and paste your HLSL code that allegedly doesn't work under the Cg compiler. You just run your mouth expecting hot air to subsitute for reasoned analysis. Let me give you a little hint: it won't.

"but doing what a tabloid would ... i mean christ if anyone one wants a better example your doing this all they have to do is read the post you've taken your last bunch of quotes from and see that how you've spliced up what i've said was taken to context your arugment and not what was really said."

And where did I misrepresent you? At what point did I distort your argument(and I use that word lightly)? Now it can't be too hard for you to cut and paste the section(or sections) of that last post to actually back up your claim, can it?

"unlike you i'm not sitting here trying to bullshit my way to win an arugment,"

And what do you call your repeated refusal to offer any evidence in favor of your side? If thats not bullsh*ting then I don't know what is.

"this is why i rarely if ever use quotes to further my point - because it is upto people to read the information for themselves and come across what is said"

Isn't that what I'm doing? You seem to have a big problem with me doing this but its okay for other people?

If you really believed this then why do you post at all? If what you say is true then you wouldn't and shouldn't post anything helpful at all because that would be preventing people from looking up the information themselves. If anything the only thing you would post is a link to the info and that would be that.

"I could quote your 12ways from Sunday and come up with valid arguments against you as well,"

In a pig's eye you could.

"but i don't need to becaus although you can blag what you know and fool people as i said you can't fool the compiler (that is if you even use the compiler)"

But you can apparently fool yourself pretty well what with this phantom HLSL shader that won't compile "first time" and all.

"the way i see it here all you want to do is continue to argue your point,"

How perceptive of you.

" even once i've said "fine believe what you want" which as Rob mentioned would really show you've won the argument"

An admission of defeat? From you?! Who are you and what have you done with the real Raven.

" you still wanted to push for further conflict."

And you are different from me in this how?

"if you really believed what you were saying then you wouldn't have posted again its as simple as that."

So let me get this straight. If I really believed what I was saying then I wouldn't argue with you. Is that it? If this is true then that must mean that you don't really believe what you are saying because you are continueing to argue with me as well, right? That would be the logic of it correct?

Let me get things clear with you. I believe that HLSL is basically identical to Cg and that what will compile in one will almost always compile with the other. Now I have supported what I said with numerous examples and even explained one of the small differences between Cg and HLSL.

You, on the other hand, continue to insist that they are somehow a lot different in the face of every shred of evidence to the contrary. If you want to be stubborn and not admit that you are wrong then fine. But don't ever make out like it is I that is the ignorant one.

@Raven

::sigh::
Its getting late so I'll keep this short and sweet.

"Quote:

we're both moving toward that compatibility


keywords meaning that they are not right now like you claimed... i'd start reading your own quotes."

I'd start reading the links if I were you. If you check you'll see that that interview was taken during july 2002. The "gold release" that it mentions has already occured.

If you want a challenging exercise you can put those two refraction shaders into a dbpro programming and make them go I had a quick go and manage to load them onto an object but dont know if thats right. I just used the help code for shaders, but when I apply the vertex shader the object disappears and the pixel shader slows the fps down to almost nothing.

because unlike you i'm not trying to push for any conflicting objections, i really don't give a shit what you think or say anymore ... how i see it is that you still feel you need to argue a point and quite frankly i didn't even bother to read you last post, as i can't be bothered to waste the time - so have fun talking to that brick wall across from you because i'm sure if you talk long enough it'll turn around and say your right about everything.

for some reason that appears to be what your craving here and i'm not going to tell you something i know is a lie.

would like to say though, i talked to Kennith Wakoywski yesterday about the HLSL & Cg language things ... infact my exact question was "does Cg work the exact same way as HLSL?" as he is the guy who developed the Cg compiler he did seem like a sound choice to ask.
the only response i got was him laughing followed by "you know kid i don't know how you could keep a straight face askin' that." promptly followed by him handing me the new compiler on a CD.

you can take that however you like to aid whatever point you want to make, or against. really i don't care
as for expecting an example to be coded, he used aruging with me as an excuse why he didn't rewrite in ASM or use the Compiler until now ... draw your own conclusions from that.

"you know kid i don't know how you could keep a straight face askin' that"

So does anyone believe that he really said that?

[Tumbleweed]

I'm sorry Raven, but produce some better evidence than that, come on, find us a SINGLE web quote, and given the vast wealth of material available on the net, there will be something that verifies your claims if what you say is true.

Still, I'll ask Lee or Mike on this one

Found this quote as well:

"Cg is an industry standard high-level graphics language developed by NVIDIA with Microsoft's participation; it's 100% compatible with the DirectX 9 High Level Shading Language. That means that Cg code can be compiled into DirectX commands. (Compiling is the process of converting a high-level language like C into raw assembly language that is used to create a computer program executable (exe) file."

http://www.bjorn3d.com/_preview.php?articleID=22&pageID=63

http://www.nvidia.com/view.asp?IO=cg

Quote from the site

@Elemenop

"If you want a challenging exercise you can put those two refraction shaders into a dbpro programming and make them go"

Sadly, that is impossiable at the current moment. In order for those shaders to be used one would need the texture coordinates for the cubic enviroment map and there is currently no way to get those coordinates from within DBPro. That might all change with patch 5 though. If it does, and I'm allowed access to the cube map coordinates, then I'll make a working project and post it on this board. I promise.

@Raven

"because unlike you i'm not trying to push for any conflicting objections"

And what do you call this post of yours? And the alledged statement by Kennith Wakoywski at the bottom? If you weren't pushing for futher conflict then why did you respond to my Nvidia quote post just after you got done saying that you were done? You are so full of it.

"i really don't give a shit what you think or say anymore"

Did you ever?

"how i see it is that you still feel you need to argue a point"

Have you ever picked up on this recurring habit of yours of projecting your own faults on to me? I know I have.

"and quite frankly i didn't even bother to read you last post, as i can't be bothered to waste the time "

And yet you can find the time to type this post? You know I think you actually did read my post. You are just to embarassed to admit you made a total fool out of yourself by not bothering to read the link.

"so have fun talking to that brick wall across from you because i'm sure if you talk long enough it'll turn around and say your right about everything."

At what point have I not been talking to a brick wall? Every point I've made, every shred of evidence that I've put forward you either totally ignored or dismissed out of hand. Do you enjoy being so hard headed? Would it kill you to admit that you goofed even just once?

"for some reason that appears to be what your craving here and i'm not going to tell you something i know is a lie"

How do you know it is a "lie"? And which(you were wrong about a lot of things in this thread) paticular "lie" are you not willing to tell?

"would like to say though, i talked to Kennith Wakoywski yesterday about the HLSL & Cg language things ... infact my exact question was "does Cg work the exact same way as HLSL?" as he is the guy who developed the Cg compiler he did seem like a sound choice to ask."

Sure you did.

"the only response i got was him laughing followed by "you know kid i don't know how you could keep a straight face askin' that." promptly followed by him handing me the new compiler on a CD."

Is this suppose to mean something? Assuming its true, as far as I can tell this can be interpreted anyway one wants. Or are you not telling us the whole quote?

"you can take that however you like to aid whatever point you want to make, or against. really i don't care"

Am I senseing a little doubt here? Could it be that you're actually coming to the realization that you are utterly wrong about this and that Ken might actually be supporting my side of the argument? Or am I being too optimistic? I think I'm being too optimistic.

"as for expecting an example to be coded, he used aruging with me as an excuse why he didn't rewrite in ASM or use the Compiler until now ... draw your own conclusions from that."

And work. Don't forget work. I don't know about you Raven but I have a life outside these boards and at the moment it is a little demanding(you have been noticing the times I've been posting these haven't you?). As much as I would like to argue with you all day about this I have other obligations that I need fulfill. They really don't leave me with a lot of time to spend on these boards or code examples. I have no idea by what you mean by "draw other conclusions" and I don't think you do so I'll ignore it for the time being(unless of coarse you would like to argue about that.

@Rob K

"So does anyone believe that he really said that?"

Probably not, no. Raven does have a tendency to make things up.

"I'm sorry Raven, but produce some better evidence than that, come on, find us a SINGLE web quote, and given the vast wealth of material available on the net, there will be something that verifies your claims if what you say is true."

I wish I could share you're optimisum but it doesn't appear like he'll being doing that any time soon so I'm sorry to say you're wasting your time. Raven knows he has lost this argument and badly. He's just looking for a way to salvage his ego it appears. Besides, as he has stated before, backing up what he says isn't his "style."

"backing up what he says isn't his "style.""

I'm still waiting for the web link that tells me that the GBA has a 300 MHz processor

:: yawn ::
you want proof about me being right the answer is in the quotes you yourself have provided.

somehow i kinda get the feeling you are one of those people who believe that C and C++ are one and the same...

you want me to explain this shortly and sweetly how i am right here and you are wrong through a single quote, i will use this pretty topic famous one


Compatibility even 100% doesn't not mean the languages are identical, infact at no point does it say you can rename a .cg as .fx and use it as such.
at no point does it say the HLSL has the same ammount of commands, structures and input/outputs as Cg.
at no point does it say that Cg can use the DirectX HLSL declarations in standard format...
nor does it explain why the APi for DirectX & OpenGL have these commands, which if what you say is true would make them a complete waste of time



(and that is just on the minimum interface)
you want proof it is staring you right in the face, languages can be 100% compatible, they can use the same syntax and still be COMPLETELY DIFFERENT languages.
Pascal & Basic would be another very very good example.

only difference is as not to seperate themselves from the industry standard DirectX they've decided to make them with interchangeable parts, just like Pascal and C have.

perhaps now it was time you stoped your epic posts and were quiet about a subject you obviously know NOTHING about.

Click this link http://www.google.com/ and find a site that backs up what you are saying or proves the other view wrong.

Then stop blabbing on about nonsense like this.


Which isn't even true, basic and pascal do not share syntax and are not 100% compatable.



Stop clutching at straws and stick to the original argument. No one was saying that Cg and HLSL where implemented or work in the same way just that the code is compatable. This view is expressed on the nvidia website and others. The links have already been posted.

"somehow i kinda get the feeling you are one of those people who believe that C and C++ are one and the same..."

I don't believe that and I can find plenty of evidence to explain otherwise. There is a whole chapter devoted to differences between C and C++ in my tut. book - does the same thing exist for HLSL / Cg, my ass it does.

"Pascal & Basic would be another very very good example."

Stop making comparisons and actually back up your argument for once dammit. Pascal and Basic do not share identical syntax and are not compatible. Delphi will not compile Basic, VB will not compile pascal. Yet HLSL will compile in a Cg compiler and vice versa according to the quote you posted.

Basic:



Pascal:



As you can see, Pascal is not basic, it is in fact a much closer relative of C.

"Compatibility even 100% doesn't not mean the languages are identical, infact at no point does it say you can rename a .cg as .fx and use it as such.
at no point does it say the HLSL has the same ammount of commands, structures and input/outputs as Cg.
at no point does it say that Cg can use the DirectX HLSL declarations in standard format...
nor does it explain why the APi for DirectX & OpenGL have these commands, which if what you say is true would make them a complete waste of time"

However the quote infers all of those things, thanks for prooving Neo's point

At the end of the day, you can pussyfoot around as much as you like, but you haven't even come up with a SINGLE quote OF YOUR OWN which prooves your point. Apart from that one slightly vague quote, there is a lot of evidence that Cg and HLSL are just alternatives for which developers can pick the one that suits them.

Can you

a) Restate clearly, explicitly and in detail (not general terms) what the differences ARE in your opinion.

b) Back this up with legit. webquotes + links to the relevant webpages?

"perhaps now it was time you stoped your epic posts and were quiet about a subject you obviously know NOTHING about."

And Raven... the GBA STILL doesn't have a 300Mhz processor.

tell ya what, you guy win ... i'll leave shader dictating to the obvious pro's in this field.

"obvious pro's in this field"

Should read "pros"

@Rob K

"I'm still waiting for the web link that tells me that the GBA has a 300 MHz processor "



He actually said that? Where? I've got to see this.

@Raven

"tell ya what, you guy win ... i'll leave shader dictating to the obvious pro's in this field."

Thank you.

*cough* http://www.darkbasicpro.com/apollo/view.php?t=10819&b=2 *cough*

Sorry about this Raven, but you asked for it.