It is a *very* good thing that NVidia have got some competition. While NVidia are a great company, from my perspective as a developer, the fact that they are coming close to ruling every market unsettles me slightly.
Radeon I was a bit of a disappointment as far as I could make out, not quite cheap enough to be a budget card but not quite good enough to take on GF2. The 8500 looks to be quite a nice piece of kit, and although I wasn't sure at first, the extended Pixel Shader caps should be very good fun to play with.
However, the current benchmarks don't put the 8500 far enough ahead of the GF3 for it to be a clear win, especially since the 8500 will be about GBP350 when it arrives, and I can get hold of a GF3 for under GBP250. What matters to ATI is the driver support - they need to get good enough drivers out of the door to put a clear gap between them and the GF3 in terms of performance, and plenty of decent developer relations to emphasise the feature set (although TruForm doesn't excite me at all - look ma! Hardware tesselation *all the time*!). Otherwise, NVidia will release their next part which will trounce the 8500 (don't imagine it's far away), before ATI have had a chance to reclaim their market share.
I wonder exactly what market ATI are aiming at - will the hardcore gamer market really offer them high enough sales to make a comeback? Or will they target the OEM market, where they used to be king?
Fair enough - I did miss the point. I didn't realise that Microsoft were in charge of signing the code. This could obviously lead to a sword of Damocles type situation where Microsoft are, like you suggest, making unreasonable requirements of the drivers. Thanks for putting me straight.
Henry
Some of this is just stupid....
on
Dan Gillmor on WinXP
·
· Score: 5, Insightful
I don't want to come across as one of the "everything/. posts about MS is biased" crowd, so I won't comment on the decision to post this. However, some of the points in the article made me laugh.
Microsoft has added ''code-signing'' measures -- verification, supposedly, that downloads will be safe -- that could scare customers away from using software that competes with Microsoft's offerings.
How do you spell FUD? This is just silly. Microsoft have added code-signing (which I thought had been around a while) - which they could use to scare people away? How? I suppose they could do something by only allowing MS code to get signed or something, but that's pretty damn unlikely. The idea is that you can be certain where the code has come from, and then it is up to you to decide whether you trust it. Microsoft add *no* commentary on whether they think you should trust it or not, and to assume they will do is just paranoia.
Microsoft removed the Java environment from XP, thereby breaking thousands of Web sites that use Java. XP customers will face endless downloads to replace the functionality they'd come to expect.
While I'd rather the JVM was still in XP (but I hadn't noticed it was gone, will check that tonight when I get home), I don't feel any anger towards Microsoft for removing it. They have a competing platform, in.NET. Their attempts to do something with Java, rightly or wrongly, resulted in them getting their wrists slapped. No-one at MS that I've talked to really cares that much about Java. So why should they include it?
Microsoft is bundling all kinds of services into XP in ways that block competition, from photography software to video/audio playback. If customers want to use other vendors' products they'll have to jump through Microsoft-designed hoops
Slightly more questionable this. But I do like having ZIP folders natively as part of Explorer. But I've never had any problems with replacing functionality with the alternatives. I am an *informed* user. It is my business as a user to remain informed, and to make the choices that are right for me.
I could go on, and the article makes points about the OEM market that do sound pretty worrying to me. But all this article does is regurgitate some of the common fears and rumours surrounding XP, without *any* real and substantial justification of this strange injunction idea. I agree with authentication of XP, since you can control what is being posted, and I don't think piracy is good. I haven't had to use my Passport once, and I've been using XP since Beta 1 (as in never - have never even typed in the password).
I guess I just don't understand why people are making such a noise about fairly minor complaints. My cynical side is telling me that it's indicative of a jealousy of success, but I don't think that's always the case.
Game Institute offers courses (non-trivial ones at that) for the aspiring game maker. They look pretty good, and I've heard good things about them, but I haven't done one myself.
The thing about game development is that is rapidly turning into its own kind of engineering. Large projects neccesitate good engineering practice. However, there is reputedly still remarkable reluctance on the part of developers to adopt coding practices that have been the norm in other development fields for some time (the adoption of C++ for one, but I realise that can start a flame war, so don't).
I don't think it's reasonable to say that game development is an academic discipline - a reasonable acid test is whether there is active research in game development. There's loads in graphics and visualisation, probably a bundle in audio techniques, and a lot of AI... but these are all ends in themselves, rather than explicitly contributory research to the field. Most implementations of research techniques are very heavily tailored due to the contraints placed upon the games developers by technology.
That's not to say that game-development doesn't take skill - clearly there are some incredibly bright people working in the field. It certainly warrants its own journal. Maybe we'll see some standardisation bodies:)
This would be done more through code than by an artist
True, but what you go on to say is that the Phong constants are indeed constant across a surface - then ATI saying 'oh look - you can programatically change ka and ks' becomes useless because you won't need to change it. This assumes that you are working on a one material : one texture map correspondance. If, like their examples, you have say metal and stone on one map then varying some constants becomes necessary. But then this requires another map (or even two) at close to the resolution of the source diffuse map. You can do per-material ka/kd/ks now with no troubles at all. Per-fragment is a bit more involved.
Why not? They draw textures now,
I'm not an artist, so I don't know. But I don't know that the tools are there for them to draw bump-maps, and you have to admit that using an RGB channel as a three component normal vector can't be the most intuitive way to draw things. Much better to procedurally generate, like you say.
Stanford have been doing some research on compiling RenderMan (pretty much the holy grail) shaders down to OpenGL. They can do it, in a lot of passes and with a couple of extensions.
I have heard that there is work ahead to do this with Pixel Shaders. Once Pixel Shaders become sufficiently general, all you need to do is re-target the back end of the compiler and you're set.
But now we have API standards. Even better, I think. Graphics card manufacturers have complete freedom of implementation. It's been a while since games actually didn't work on one of the big cards (and quite a few still work on Voodoo 3s).
I love things working right out of the box, it's one of the reasons I still use Win2k/XP for almost everything. I agree, standardisation of hardware would bring this kind of compatibility within the reach of others. But it would probably lead to a lowest common denominator approach, and then somebody will ignore the specs. It'll all happen again.
Say what? NVidia's cards have always rocked (except the ZX chipset admittedly), I agree. But NVidia provide a level of community support *far and away* better than ATI. NVidia host conferences for grad students and their professors. They have developer conferences in many different countries. Matt and Cass from NVidia hang out on opengl.org's discussion forums and help everyone out (newbies, old hands, the lot). The developer documentation is sublime - and everyone can get at it. Plus their drivers *just work* 9 times out of 10.
I could care less about driver specs. The 3dfx ones are around if I want to see how modern-ish graphics cards are set up. And their drivers are such good quality, I can see why they don't want mutations springing up all over the web. I certainly don't have a problem with such a pleasant company to work with wanting to hold on to a few secrets.
First of all, a direct link to ATI's SmartShader tech introduction.
I have a few disparate thoughts on this subject, but rather than scatter them throughout the messages I'll put 'em all in one place.
ATI are attacking what is possibly the weakest part IMHO of DirectX 8 - the pixel shaders. Pixel shaders operate on the per-fragment level, rather than on the per-vertex level vertex shaders which were actually Quite Good. The problem with Pixel Shaders 1.1 is that, to paraphrase John Carmack, "You can't just do a bunch of math and then an arbitary texture read" - the instruction set seemed to be tailored towards enabling a few (cool) effects, rather than supplying a generic framework. Again, to quote Carmack, "It's like EMBM writ large". Read a recent.plan of his if you want to read more.
If you read the ATI paper, they don't really tell you what they've done - just a lot of promises, and a couple of "more flexibles!", "more better!" kind of lip-service. I don't care about reducing the pass number. Hardware is getting faster. True per-pixel phong shading looks nice, but then all they seem to do extra is allow you to vary some constants across the object via texture addresses. Well that's great, but texture upload bandwidth is can already be significant bottleneck, so I don't know for sure that artists are gonna be able to create and leverage a separate ka, ks etc map for each material. (I did enjoy their attempts to make Phong's equation look as difficult as possible)
True bump-mapping? NVidia do a very good looking bump-map. Adding multiple bump-maps is very definitely an obvious evolutionary step, but again, producing the tools for these things is going to be key. Artists won't draw bump-maps.
Their hair model looks like crap. Sorry, but even as a simple anisotropic reflection example (which again NVidia have had papers on for ages) it looks like ass. Procedural textures, though, are cool - these will save on texture uploads if they're done right.
What does worry me is that the whole idea of getting NVidia and Microsoft together to do Pixel Shaders and Vertex Shaders is so that the instruction set would be universally adopted. Unfortunately, ATI seem to have said "Sod that, we'll wait for Pixel Shader 1.4 (or whatever) and support that." I hope that doesn't come back to bite them. DirectX 8.0 games are few and far between at the moment, so when they do come out there'll be a period when only Nvidia's cards will really cut it (I don't think ATI have a PS 1.0 implementation, someone please correct me if I'm wrong) - will skipping a generation hurt ATI, given that they're losing the OEM market share as well?
I dunno, this just seems like a lot of hype, little content.
.NET is a whole load of technologies, the CLR is only one of them. I don't know if a CLR port will necessarily be able to run:
Visual Studio.NET
ASP.NET
Any of the enterprise servers (Commerce, ISA, Exchange, BizTalk, SQL Server)
These technologies are where the real use of.NET lies, not just the technical interest.
If a CLR port can run these products, then even better for Microsoft, since it can run its server products on every operating system. But I think they'll make calls down to System32 which won't be portable.
I'm posting from Microsoft's TechEd conference in Barcelona, where as you might expect,.NET and its associated technologies are the focus of many presentations (not all though - been to some wicked ones on Win2k / XP internals).
Let me try and put some things straight, since I've seen many a post asking 'WTF is.NET?' and have yet to see a good answer.
.NET is the marketing rename of Windows DNA which was conceived in the mid to late nineties. It is an umbrella term for a whole host of technologies, some of which are cool, some of which are just ok.
The Common Language Runtime is pretty self-explanatory. The CLR provides a run-time environment for programs compiled specifically for it. You can draw a very close analogy between the JavaVM and the CLR. CLR differs in a couple of places: There is no bytecode interpreter in the CLR, everything is JIT compiled either at installation or first-run. Also, CLR is not language specific (JavaVM isn't really either, but...) - there are many languages that compile to the intermediate code, and at that point they are practically indistinguishable.
C# is Microsoft's Java-a-like which works on their platform. It's very similar syntactically (and in the class heirachy) to Java, but has a couple of evolutionary features - properties for instance, and attributes. No one is claiming that C# is a revolutionary language, but what it is is standardised, open and pleasant to use (of course YMMV). C# isn't really the focus of.NET, due to the language independance of the CLR, but it's a quality supporting player.
Simple Object Access Protocol (SOAP - combine with a Remote Object access protocol to get SOAP on a ROAP:) is a Microsoft and partners technology to do better RPC. SOAP takes advantage of HTTP and XML to transmit procedure calls across a network (a couple example advantages: good compression, goes through firewalls on the HTTP port). If you get the chance, read some of Don Box's articles in MSDN, since he was heavily involved (and see his presentations - although he may very well be naked:) SOAP is obviously entirely platform independent (can do HTTP? can do SOAP.)
SOAP provides the foundation for Web Services, marketing speak for web-based applications that take advantage of XML for data communication. The vast majority of web services that I've seen this past week are client-independent (except for requiring DHTML support quite often). The changes have come on the server side, with ASP.NET, ADO.NET and plenty other technology updates.
So there's a lot of significant updates to Microsoft technologies, and the introduction of some new ones, which together present a platform that doesn't lock everyone into Microsoft specific products, unlike the FUD here sometimes suggests. XML is open, SOAP is open - these are the communication protocols. The CLR is open, C# is open. If CLR got ported around it would be a great thing - because the some of the ideas that Microsoft is pushing are actually quite good things IMO. Visual Studio.NET isn't open - well what did you expect? Microsoft's business model relies on revenue from product sales of these tools. But it is a very good set of tools.
The article seems a little contradictory, saying that the Linux desktop is entirely dead, but still saying that enthusiasts will continue to work on it. To my mind, continued work on something implies that it's probably not dead.
However, I agree with a lot of the rest of the article. Linux is a superb server platform as far as I can tell. I don't know about truly industrial strength applications since I've never used it that way, but just for setting up a web server, firewall, mail server, QuakeWorld server it rocks. But for desktop productivity... eww. The window managers are slow and clunky (Sawfish is the best that I've found, but that takes an age to start), most of the applications are flakey in a random way that I don't see on Windows - right now the Gnome File Manager keeps telling me that there's no response to a save yourself command, and would I like to remove it? Clicking yes does absolutely nothing. Sure loads of Windows applications crash, but they crash, and stay dead. I can get rid of them. (even kill -9 doesn't seem to work properly). And don't even get me started on the productivity software...
It all comes down to using the right tools for the job. Linux does a great deal of server work for me in the networks I am involved with, Windows does a lot of the desktop work. Even if the Linux desktop is dead (and I don't think it is, it just needs improving) that's not really that big a deal for Linux.
It's already done, to lesser or greater extents. From the time of Quake (and before), many different techniques are used to cut the amount of geometry being processed. With graphics cards as agood as they are now, it's still an issue (maximise bandwidth usage, fillrate etc). A lot of graphics work is an exercise in culling away work at different stages of the pipeline. The key is to do this as early as possible, to maximise the benefits.
Quake makes use of a PVS (Potentially Visible Set), which does just what you describe, making a conservative estimate of the visible geometry at any time, and only drawing that. Fluid Studios have a technique to make the estimate near (completely?) exact, but clearly they're not sharing too many details at this point. And of course there's simple frustum culling, which gets rid of objects outside of the field of view. And....
With open source drivers (by which I mean the source freely available, independant of license), wouldn't this debate then become academic? Since anyone could have access to the source to produce drivers, this feature will make it into somebody's code base somewhere. This would essentially reduce to another Quake-style cheating thing since clients could be using altered binaries without the server's knowledge. Possibly an argument against releasing driver source? Who knows...
The fact that ASUS are touting these as drivers for cheating (rather than just saying that the see-through feature is a 'special effect' or somesuch) leaves a bad taste. Although gaming clearly isn't the most critical application of computing, it doesn't absolve companies of responsibility towards their customers.
Nvidia's website has some demos on how to do sketch rendering on their hardware. Intel have done some pretty cool research on sketch rendering and cartoon rendering.
This page has a lot of good links. Check any recent Siggraph set of proceedings as well.
Most current techniques seem to involve "thresholding" the Lambertian diffuse lighting equation, so banding the colours used, or using that as in index into a 1d texture map which contains the different shades used for a model.
Hmm. Check the websites above for a better explanation than that:)
I don't claim to be enlightened, but I'll give it a go.
CPU seems a little underspecced, but that depends on the amount of off-chip processing available. For comparison purposes, the X-BOX will ship with a 700MHz processor, which won't touch the graphics pipeline at all.
Reasonable amount of memory. Be interesting to see if they adopt a unified memory architecture (again, cf. X-BOX).
Integrated video chip is definately the norm for consoles - no expansion required, therefore reduce costs and optimise data paths by putting everything on the same board. The feature list was bog-standard a couple of years ago around the time of the TNT(2). It's only a little up on the original PlayStation. I speculate that they may not be pushing this console as a direct contender to X-BOX etc. unless they have some pretty significant improvements to make graphically, since that is where, to an extent, the war is being fought at the moment. Current hardware is touting 'fully' programmable pipelines, along with a whole lot of cool features (cube environement mapping, per-pixel lighting, shadow generation in hardware, bump-mapping etc).
This would be a capable 3d machine a year or two ago. Now it could be considered a little dated.
The graphics card area of interest has moved away somewhat from the super-high fill rate battles that were all the rage in the day of the Voodoo cards.
It's all about interesting and orthogonal features now. GeForce 3 brings vertex and pixel shaders in hardware to the mix, as well as hardware shadow map support. The disappointing thing is that 3D textures (despite word otherwise from John Carmack) don't appear to be accelerated in hardware, at least with latest drivers (see a recent thread in the advanced section of www.opengl.org on that unfolding story - NVidia are soon to make an announcement on what the deal really is).
Being able to program in a pseudo-assembler language for custom per-pixel effects is a hark back to the old days when you had complete freedom over everything you could do, but most of it was slow. Now we have a better mix where we are hardware accelerated, but pretty flexible down to a programmable level. *However* the current revision of pixel shaders (1.0? 1.1? can't remember) on DirectX (and very similarly and more relevantly on OpenGL) aren't as flexible as some may like (notably John Carmack), since to paraphrase him "You can't just do a bunch of maths and lookup a texture". Hopefully that will get better with time.
Yes these things are important to games mostly. And yes they are arguably the biggest step forward in consumer graphics tech since the original 3dfx card... certainly since hardware TnL. Wait for the price to come down (since initial pricing is aimed at developers and the *really* hardcore gamer), and in the meantime amuse yourself with some of the demos from Nvidia's developer site. Nvidia are by far the most developer friendly company I've ever encountered, so short of Open Sourcing their drivers (which we have no right to expect them to do), they are almost ideal from my (developer's) perspective.
Henry
Re:A worrying turn of phrase...
on
Paper Phones
·
· Score: 1
No that's not what I'm saying. Read the quote. I ascribe the same responsibility to society itself.
However, I think that what we have freely available should be controlled to some extent, and it already is. Cars, guns, drugs, houses... all controlled to some extent by some organisation.
I would just like to see those people responsible for bringing abusable products to market take some sort of responsibility for their use, and take reasonable steps to prevent their abuse. Is that so unreasonable and freedom-removing?
Henry
Re:A worrying turn of phrase...
on
Paper Phones
·
· Score: 1
You miss the point.
She shouldn't seek to control how her invention is used - but she should perhaps be aware of possible misuses, and her part in allowing them to be perpetrated. I don't expect her to remove the product, or take other drastic measures to combat the litter issue. However, I do expect her to judge morally and ethically what she is bringing to the world, and not absolve herself of any involvement or responsibility by simply passing the buck to the bodyless abstract of society.
It is not a matter as simple, or as crude, as "whose fault is this". While of course, ultimate responsibility lies with the wilful misuser, should she not feel she has a duty to do as much as is reasonable to prevent the misuse?
This last question is not rhetoric, it is an interesting problem: should we allow society to govern itself (and give everybody guns), or should we limit its capabilities to harm itself by removing a litte freedom?
Therefore, no caller ID, no text messaging and most importantly, no Snake.
A phone's communicative capabilities are incidental to its game playing ones.
Henry
A worrying turn of phrase...
on
Paper Phones
·
· Score: 3
One quote disturbs me philosphically:
"I can't change what society is. We are a disposable society. Life is what it is."
I was reading one of Feynamann's "Meaning of It All" Lectures last night (specifically Uncertainty in Science). There, he talks about what, if any, the responsibility of the scientist (engineer, whatever) is to society in bringing to bear the applications of an idea. He said that each scientific idea presented the "keys to the gates of heaven, and of hell" (paraphrasing). While it would be foolish to pass up the opportunites of a key to heaven, it would be unwise to not consider the possibility of hell.
While of course, heaven and hell are exaggerations here, the principle holds. I sincerely hope this woman has not passed the buck of responsibility for potential wastefulness to society as a whole. As the sole person responsible for bringing this product to society (and only that - it seems a team of engineers were responsible for design and implementation), she has a responsibility (as clearly society does) to make sure her invention is used in a proper manner.
True, but this particular chip isn't just a faster card. In fact, SharkyExtreme (and others I've read), only expect to see large improvements when FSAA is enabled, due to the new reconstructive filtering technology used, rather than super sampling.
What NVidia have done (and Microsoft with DX8), amongst others, is introduce progammability of the pipeline to consumer cards. This is really important stuff, and is a much bigger deal than the big things of the last few generations (hardware TnL, for instance, hasn't had the huge impact that was anticipated). We're getting closer to real-time RenderMan. (There's a paper by some guy at SGI saying that with current generation cards (and a couple extensions), RenderMan shaders can be implemented. Just not very quickly).
John Carmack mentioned in his.plan that this new card had enough new important features to make graphics programmers very happy indeed. I agree with his conclusion that pixel-shaders aren't general enough yet, but they can still do some cool stuff.
in favour of the Honour System. Visitors get the chance to donate an arbitary amount of money.
Penny Arcade are doing this, and are seeing quite a substantial sum donated. Don't know how this compares to banner ads... did anyone ever click on them anyway?
Time will tell whether the honour system actually works well enough for people to make a living off of it.
Yeah, I do care. I visit msdn.microsoft.com all the time, because it's full of interesting technical information (not just for microsoft platforms), which appeals to the geek in me.
As a non-politically driven techie, I find I'm able to swallow my pride and make best use of all the information available. I don't feel somehow dirty after visiting ms' site.
While I understand a large part of slashdot's readership are none too big fans of microsoft's, I hope no-one attempts to generalise Katz-style about what we _all_ are interested in. I appreciate finding the line to tread is difficult.
What is slightly more interesting is the motivation behind people submitting the story. I suppose people wanted to laugh at ms' technology because it went wrong, and garner some reflected sense of superiority from that. Hmm.
Slashdot Mirror
It is a *very* good thing that NVidia have got some competition. While NVidia are a great company, from my perspective as a developer, the fact that they are coming close to ruling every market unsettles me slightly.
Radeon I was a bit of a disappointment as far as I could make out, not quite cheap enough to be a budget card but not quite good enough to take on GF2. The 8500 looks to be quite a nice piece of kit, and although I wasn't sure at first, the extended Pixel Shader caps should be very good fun to play with.
However, the current benchmarks don't put the 8500 far enough ahead of the GF3 for it to be a clear win, especially since the 8500 will be about GBP350 when it arrives, and I can get hold of a GF3 for under GBP250. What matters to ATI is the driver support - they need to get good enough drivers out of the door to put a clear gap between them and the GF3 in terms of performance, and plenty of decent developer relations to emphasise the feature set (although TruForm doesn't excite me at all - look ma! Hardware tesselation *all the time*!). Otherwise, NVidia will release their next part which will trounce the 8500 (don't imagine it's far away), before ATI have had a chance to reclaim their market share.
I wonder exactly what market ATI are aiming at - will the hardcore gamer market really offer them high enough sales to make a comeback? Or will they target the OEM market, where they used to be king?
Interesting times.
Henry
You can skin XP right now. Unfortunately, when trying to look for more themes, there were none on the website I was directed to.
Henry
Fair enough - I did miss the point. I didn't realise that Microsoft were in charge of signing the code. This could obviously lead to a sword of Damocles type situation where Microsoft are, like you suggest, making unreasonable requirements of the drivers. Thanks for putting me straight.
Henry
I don't want to come across as one of the "everything /. posts about MS is biased" crowd, so I won't comment on the decision to post this. However, some of the points in the article made me laugh.
.NET. Their attempts to do something with Java, rightly or wrongly, resulted in them getting their wrists slapped. No-one at MS that I've talked to really cares that much about Java. So why should they include it?
Microsoft has added ''code-signing'' measures -- verification, supposedly, that downloads will be safe -- that could scare customers away from using software that competes with Microsoft's offerings.
How do you spell FUD? This is just silly. Microsoft have added code-signing (which I thought had been around a while) - which they could use to scare people away? How? I suppose they could do something by only allowing MS code to get signed or something, but that's pretty damn unlikely. The idea is that you can be certain where the code has come from, and then it is up to you to decide whether you trust it. Microsoft add *no* commentary on whether they think you should trust it or not, and to assume they will do is just paranoia.
Microsoft removed the Java environment from XP, thereby breaking thousands of Web sites that use Java. XP customers will face endless downloads to replace the functionality they'd come to expect.
While I'd rather the JVM was still in XP (but I hadn't noticed it was gone, will check that tonight when I get home), I don't feel any anger towards Microsoft for removing it. They have a competing platform, in
Microsoft is bundling all kinds of services into XP in ways that block competition, from photography software to video/audio playback. If customers want to use other vendors' products they'll have to jump through Microsoft-designed hoops
Slightly more questionable this. But I do like having ZIP folders natively as part of Explorer. But I've never had any problems with replacing functionality with the alternatives. I am an *informed* user. It is my business as a user to remain informed, and to make the choices that are right for me.
I could go on, and the article makes points about the OEM market that do sound pretty worrying to me. But all this article does is regurgitate some of the common fears and rumours surrounding XP, without *any* real and substantial justification of this strange injunction idea. I agree with authentication of XP, since you can control what is being posted, and I don't think piracy is good. I haven't had to use my Passport once, and I've been using XP since Beta 1 (as in never - have never even typed in the password).
I guess I just don't understand why people are making such a noise about fairly minor complaints. My cynical side is telling me that it's indicative of a jealousy of success, but I don't think that's always the case.
Henry
Game Institute offers courses (non-trivial ones at that) for the aspiring game maker. They look pretty good, and I've heard good things about them, but I haven't done one myself.
:)
The thing about game development is that is rapidly turning into its own kind of engineering. Large projects neccesitate good engineering practice. However, there is reputedly still remarkable reluctance on the part of developers to adopt coding practices that have been the norm in other development fields for some time (the adoption of C++ for one, but I realise that can start a flame war, so don't).
I don't think it's reasonable to say that game development is an academic discipline - a reasonable acid test is whether there is active research in game development. There's loads in graphics and visualisation, probably a bundle in audio techniques, and a lot of AI... but these are all ends in themselves, rather than explicitly contributory research to the field. Most implementations of research techniques are very heavily tailored due to the contraints placed upon the games developers by technology.
That's not to say that game-development doesn't take skill - clearly there are some incredibly bright people working in the field. It certainly warrants its own journal. Maybe we'll see some standardisation bodies
Henry
This would be done more through code than by an artist
True, but what you go on to say is that the Phong constants are indeed constant across a surface - then ATI saying 'oh look - you can programatically change ka and ks' becomes useless because you won't need to change it. This assumes that you are working on a one material : one texture map correspondance. If, like their examples, you have say metal and stone on one map then varying some constants becomes necessary. But then this requires another map (or even two) at close to the resolution of the source diffuse map. You can do per-material ka/kd/ks now with no troubles at all. Per-fragment is a bit more involved.
Why not? They draw textures now,
I'm not an artist, so I don't know. But I don't know that the tools are there for them to draw bump-maps, and you have to admit that using an RGB channel as a three component normal vector can't be the most intuitive way to draw things. Much better to procedurally generate, like you say.
Henry
Stanford have been doing some research on compiling RenderMan (pretty much the holy grail) shaders down to OpenGL. They can do it, in a lot of passes and with a couple of extensions.
I have heard that there is work ahead to do this with Pixel Shaders. Once Pixel Shaders become sufficiently general, all you need to do is re-target the back end of the compiler and you're set.
Henry
But now we have API standards. Even better, I think. Graphics card manufacturers have complete freedom of implementation. It's been a while since games actually didn't work on one of the big cards (and quite a few still work on Voodoo 3s).
I love things working right out of the box, it's one of the reasons I still use Win2k/XP for almost everything. I agree, standardisation of hardware would bring this kind of compatibility within the reach of others. But it would probably lead to a lowest common denominator approach, and then somebody will ignore the specs. It'll all happen again.
Henry
Say what? NVidia's cards have always rocked (except the ZX chipset admittedly), I agree. But NVidia provide a level of community support *far and away* better than ATI. NVidia host conferences for grad students and their professors. They have developer conferences in many different countries. Matt and Cass from NVidia hang out on opengl.org's discussion forums and help everyone out (newbies, old hands, the lot). The developer documentation is sublime - and everyone can get at it. Plus their drivers *just work* 9 times out of 10.
I could care less about driver specs. The 3dfx ones are around if I want to see how modern-ish graphics cards are set up. And their drivers are such good quality, I can see why they don't want mutations springing up all over the web. I certainly don't have a problem with such a pleasant company to work with wanting to hold on to a few secrets.
Henry
First of all, a direct link to ATI's SmartShader tech introduction.
.plan of his if you want to read more.
I have a few disparate thoughts on this subject, but rather than scatter them throughout the messages I'll put 'em all in one place.
ATI are attacking what is possibly the weakest part IMHO of DirectX 8 - the pixel shaders. Pixel shaders operate on the per-fragment level, rather than on the per-vertex level vertex shaders which were actually Quite Good. The problem with Pixel Shaders 1.1 is that, to paraphrase John Carmack, "You can't just do a bunch of math and then an arbitary texture read" - the instruction set seemed to be tailored towards enabling a few (cool) effects, rather than supplying a generic framework. Again, to quote Carmack, "It's like EMBM writ large". Read a recent
If you read the ATI paper, they don't really tell you what they've done - just a lot of promises, and a couple of "more flexibles!", "more better!" kind of lip-service. I don't care about reducing the pass number. Hardware is getting faster. True per-pixel phong shading looks nice, but then all they seem to do extra is allow you to vary some constants across the object via texture addresses. Well that's great, but texture upload bandwidth is can already be significant bottleneck, so I don't know for sure that artists are gonna be able to create and leverage a separate ka, ks etc map for each material. (I did enjoy their attempts to make Phong's equation look as difficult as possible)
True bump-mapping? NVidia do a very good looking bump-map. Adding multiple bump-maps is very definitely an obvious evolutionary step, but again, producing the tools for these things is going to be key. Artists won't draw bump-maps.
Their hair model looks like crap. Sorry, but even as a simple anisotropic reflection example (which again NVidia have had papers on for ages) it looks like ass. Procedural textures, though, are cool - these will save on texture uploads if they're done right.
What does worry me is that the whole idea of getting NVidia and Microsoft together to do Pixel Shaders and Vertex Shaders is so that the instruction set would be universally adopted. Unfortunately, ATI seem to have said "Sod that, we'll wait for Pixel Shader 1.4 (or whatever) and support that." I hope that doesn't come back to bite them. DirectX 8.0 games are few and far between at the moment, so when they do come out there'll be a period when only Nvidia's cards will really cut it (I don't think ATI have a PS 1.0 implementation, someone please correct me if I'm wrong) - will skipping a generation hurt ATI, given that they're losing the OEM market share as well?
I dunno, this just seems like a lot of hype, little content.
Henry
.NET is a whole load of technologies, the CLR is only one of them. I don't know if a CLR port will necessarily be able to run:
.NET
.NET lies, not just the technical interest.
Visual Studio
ASP.NET
Any of the enterprise servers (Commerce, ISA, Exchange, BizTalk, SQL Server)
These technologies are where the real use of
If a CLR port can run these products, then even better for Microsoft, since it can run its server products on every operating system. But I think they'll make calls down to System32 which won't be portable.
Henry
I'm posting from Microsoft's TechEd conference in Barcelona, where as you might expect, .NET and its associated technologies are the focus of many presentations (not all though - been to some wicked ones on Win2k / XP internals).
.NET?' and have yet to see a good answer.
.NET is the marketing rename of Windows DNA which was conceived in the mid to late nineties. It is an umbrella term for a whole host of technologies, some of which are cool, some of which are just ok.
.NET, due to the language independance of the CLR, but it's a quality supporting player.
:) is a Microsoft and partners technology to do better RPC. SOAP takes advantage of HTTP and XML to transmit procedure calls across a network (a couple example advantages: good compression, goes through firewalls on the HTTP port). If you get the chance, read some of Don Box's articles in MSDN, since he was heavily involved (and see his presentations - although he may very well be naked :) SOAP is obviously entirely platform independent (can do HTTP? can do SOAP.)
.NET isn't open - well what did you expect? Microsoft's business model relies on revenue from product sales of these tools. But it is a very good set of tools.
Let me try and put some things straight, since I've seen many a post asking 'WTF is
The Common Language Runtime is pretty self-explanatory. The CLR provides a run-time environment for programs compiled specifically for it. You can draw a very close analogy between the JavaVM and the CLR. CLR differs in a couple of places: There is no bytecode interpreter in the CLR, everything is JIT compiled either at installation or first-run. Also, CLR is not language specific (JavaVM isn't really either, but...) - there are many languages that compile to the intermediate code, and at that point they are practically indistinguishable.
C# is Microsoft's Java-a-like which works on their platform. It's very similar syntactically (and in the class heirachy) to Java, but has a couple of evolutionary features - properties for instance, and attributes. No one is claiming that C# is a revolutionary language, but what it is is standardised, open and pleasant to use (of course YMMV). C# isn't really the focus of
Simple Object Access Protocol (SOAP - combine with a Remote Object access protocol to get SOAP on a ROAP
SOAP provides the foundation for Web Services, marketing speak for web-based applications that take advantage of XML for data communication. The vast majority of web services that I've seen this past week are client-independent (except for requiring DHTML support quite often). The changes have come on the server side, with ASP.NET, ADO.NET and plenty other technology updates.
So there's a lot of significant updates to Microsoft technologies, and the introduction of some new ones, which together present a platform that doesn't lock everyone into Microsoft specific products, unlike the FUD here sometimes suggests. XML is open, SOAP is open - these are the communication protocols. The CLR is open, C# is open. If CLR got ported around it would be a great thing - because the some of the ideas that Microsoft is pushing are actually quite good things IMO. Visual Studio
Henry
The article seems a little contradictory, saying that the Linux desktop is entirely dead, but still saying that enthusiasts will continue to work on it. To my mind, continued work on something implies that it's probably not dead.
However, I agree with a lot of the rest of the article. Linux is a superb server platform as far as I can tell. I don't know about truly industrial strength applications since I've never used it that way, but just for setting up a web server, firewall, mail server, QuakeWorld server it rocks. But for desktop productivity... eww. The window managers are slow and clunky (Sawfish is the best that I've found, but that takes an age to start), most of the applications are flakey in a random way that I don't see on Windows - right now the Gnome File Manager keeps telling me that there's no response to a save yourself command, and would I like to remove it? Clicking yes does absolutely nothing. Sure loads of Windows applications crash, but they crash, and stay dead. I can get rid of them. (even kill -9 doesn't seem to work properly). And don't even get me started on the productivity software...
It all comes down to using the right tools for the job. Linux does a great deal of server work for me in the networks I am involved with, Windows does a lot of the desktop work. Even if the Linux desktop is dead (and I don't think it is, it just needs improving) that's not really that big a deal for Linux.
Henry
It's already done, to lesser or greater extents. From the time of Quake (and before), many different techniques are used to cut the amount of geometry being processed. With graphics cards as agood as they are now, it's still an issue (maximise bandwidth usage, fillrate etc). A lot of graphics work is an exercise in culling away work at different stages of the pipeline. The key is to do this as early as possible, to maximise the benefits.
Quake makes use of a PVS (Potentially Visible Set), which does just what you describe, making a conservative estimate of the visible geometry at any time, and only drawing that. Fluid Studios have a technique to make the estimate near (completely?) exact, but clearly they're not sharing too many details at this point. And of course there's simple frustum culling, which gets rid of objects outside of the field of view. And....
Henry
With open source drivers (by which I mean the source freely available, independant of license), wouldn't this debate then become academic? Since anyone could have access to the source to produce drivers, this feature will make it into somebody's code base somewhere. This would essentially reduce to another Quake-style cheating thing since clients could be using altered binaries without the server's knowledge. Possibly an argument against releasing driver source? Who knows...
The fact that ASUS are touting these as drivers for cheating (rather than just saying that the see-through feature is a 'special effect' or somesuch) leaves a bad taste. Although gaming clearly isn't the most critical application of computing, it doesn't absolve companies of responsibility towards their customers.
Henry
Non Photorealistic Rendering is very cool indeed.
:)
Nvidia's website has some demos on how to do sketch rendering on their hardware. Intel have done some pretty cool research on sketch rendering and cartoon rendering.
This page has a lot of good links. Check any recent Siggraph set of proceedings as well.
Most current techniques seem to involve "thresholding" the Lambertian diffuse lighting equation, so banding the colours used, or using that as in index into a 1d texture map which contains the different shades used for a model.
Hmm. Check the websites above for a better explanation than that
Henry
I don't claim to be enlightened, but I'll give it a go.
CPU seems a little underspecced, but that depends on the amount of off-chip processing available. For comparison purposes, the X-BOX will ship with a 700MHz processor, which won't touch the graphics pipeline at all.
Reasonable amount of memory. Be interesting to see if they adopt a unified memory architecture (again, cf. X-BOX).
Integrated video chip is definately the norm for consoles - no expansion required, therefore reduce costs and optimise data paths by putting everything on the same board. The feature list was bog-standard a couple of years ago around the time of the TNT(2). It's only a little up on the original PlayStation. I speculate that they may not be pushing this console as a direct contender to X-BOX etc. unless they have some pretty significant improvements to make graphically, since that is where, to an extent, the war is being fought at the moment. Current hardware is touting 'fully' programmable pipelines, along with a whole lot of cool features (cube environement mapping, per-pixel lighting, shadow generation in hardware, bump-mapping etc).
This would be a capable 3d machine a year or two ago. Now it could be considered a little dated.
Henry
The graphics card area of interest has moved away somewhat from the super-high fill rate battles that were all the rage in the day of the Voodoo cards.
It's all about interesting and orthogonal features now. GeForce 3 brings vertex and pixel shaders in hardware to the mix, as well as hardware shadow map support. The disappointing thing is that 3D textures (despite word otherwise from John Carmack) don't appear to be accelerated in hardware, at least with latest drivers (see a recent thread in the advanced section of www.opengl.org on that unfolding story - NVidia are soon to make an announcement on what the deal really is).
Being able to program in a pseudo-assembler language for custom per-pixel effects is a hark back to the old days when you had complete freedom over everything you could do, but most of it was slow. Now we have a better mix where we are hardware accelerated, but pretty flexible down to a programmable level. *However* the current revision of pixel shaders (1.0? 1.1? can't remember) on DirectX (and very similarly and more relevantly on OpenGL) aren't as flexible as some may like (notably John Carmack), since to paraphrase him "You can't just do a bunch of maths and lookup a texture". Hopefully that will get better with time.
Yes these things are important to games mostly. And yes they are arguably the biggest step forward in consumer graphics tech since the original 3dfx card... certainly since hardware TnL. Wait for the price to come down (since initial pricing is aimed at developers and the *really* hardcore gamer), and in the meantime amuse yourself with some of the demos from Nvidia's developer site. Nvidia are by far the most developer friendly company I've ever encountered, so short of Open Sourcing their drivers (which we have no right to expect them to do), they are almost ideal from my (developer's) perspective.
Henry
No that's not what I'm saying. Read the quote. I ascribe the same responsibility to society itself.
However, I think that what we have freely available should be controlled to some extent, and it already is. Cars, guns, drugs, houses... all controlled to some extent by some organisation.
I would just like to see those people responsible for bringing abusable products to market take some sort of responsibility for their use, and take reasonable steps to prevent their abuse. Is that so unreasonable and freedom-removing?
Henry
You miss the point.
She shouldn't seek to control how her invention is used - but she should perhaps be aware of possible misuses, and her part in allowing them to be perpetrated. I don't expect her to remove the product, or take other drastic measures to combat the litter issue. However, I do expect her to judge morally and ethically what she is bringing to the world, and not absolve herself of any involvement or responsibility by simply passing the buck to the bodyless abstract of society.
It is not a matter as simple, or as crude, as "whose fault is this". While of course, ultimate responsibility lies with the wilful misuser, should she not feel she has a duty to do as much as is reasonable to prevent the misuse?
This last question is not rhetoric, it is an interesting problem: should we allow society to govern itself (and give everybody guns), or should we limit its capabilities to harm itself by removing a litte freedom?
Answers on a postcard.
Henry
There's no LCD on the shots that I can see.
Therefore, no caller ID, no text messaging and most importantly, no Snake.
A phone's communicative capabilities are incidental to its game playing ones.
Henry
One quote disturbs me philosphically:
"I can't change what society is. We are a disposable society. Life is what it is."
I was reading one of Feynamann's "Meaning of It All" Lectures last night (specifically Uncertainty in Science). There, he talks about what, if any, the responsibility of the scientist (engineer, whatever) is to society in bringing to bear the applications of an idea. He said that each scientific idea presented the "keys to the gates of heaven, and of hell" (paraphrasing). While it would be foolish to pass up the opportunites of a key to heaven, it would be unwise to not consider the possibility of hell.
While of course, heaven and hell are exaggerations here, the principle holds. I sincerely hope this woman has not passed the buck of responsibility for potential wastefulness to society as a whole. As the sole person responsible for bringing this product to society (and only that - it seems a team of engineers were responsible for design and implementation), she has a responsibility (as clearly society does) to make sure her invention is used in a proper manner.
Henry
True, but this particular chip isn't just a faster card. In fact, SharkyExtreme (and others I've read), only expect to see large improvements when FSAA is enabled, due to the new reconstructive filtering technology used, rather than super sampling.
.plan that this new card had enough new important features to make graphics programmers very happy indeed. I agree with his conclusion that pixel-shaders aren't general enough yet, but they can still do some cool stuff.
What NVidia have done (and Microsoft with DX8), amongst others, is introduce progammability of the pipeline to consumer cards. This is really important stuff, and is a much bigger deal than the big things of the last few generations (hardware TnL, for instance, hasn't had the huge impact that was anticipated). We're getting closer to real-time RenderMan. (There's a paper by some guy at SGI saying that with current generation cards (and a couple extensions), RenderMan shaders can be implemented. Just not very quickly).
John Carmack mentioned in his
Henry
in favour of the Honour System. Visitors get the chance to donate an arbitary amount of money.
Penny Arcade are doing this, and are seeing quite a substantial sum donated. Don't know how this compares to banner ads... did anyone ever click on them anyway?
Time will tell whether the honour system actually works well enough for people to make a living off of it.
Henry
Yeah, I do care. I visit msdn.microsoft.com all the time, because it's full of interesting technical information (not just for microsoft platforms), which appeals to the geek in me.
As a non-politically driven techie, I find I'm able to swallow my pride and make best use of all the information available. I don't feel somehow dirty after visiting ms' site.
While I understand a large part of slashdot's readership are none too big fans of microsoft's, I hope no-one attempts to generalise Katz-style about what we _all_ are interested in. I appreciate finding the line to tread is difficult.
What is slightly more interesting is the motivation behind people submitting the story. I suppose people wanted to laugh at ms' technology because it went wrong, and garner some reflected sense of superiority from that. Hmm.
Henry