Slashdot Mirror
TransGaming Releases Fast Software 3D Rendering
gavriels writes "TransGaming has just released SwiftShader, an ultra-fast software-only 3D renderer that supports Vertex and Pixel Shaders. SwiftShader dynamically compiles the geometry and rasterization pipelines to produce code that exactly matches the graphics features a game or application is using. Demo download and tech details can be found on their website."
Ad-posts such as this one, wouldn't be as offensive if they would just come out and say that they're an ad-post. I.e., instead of "on their website", say "on our website". And make some sort of comment that, yes, the poster does have a vested interest in the content of the post.
Here's the meat of the article for those who can't be bothered.
TransGaming's SwiftShader technology provides the world's fastest pure software 3D renderer with DirectX 9.0 class features, according to the company, including support for Pixel and Vertex Shaders. SwiftShader is built to provide the same APIs that developers are already using for their games and applications. This makes it possible to directly integrate SwiftShader into applications without any changes to source code. Direct3D 8 and Direct3D 9 compatible APIs are available immediately, and OpenGL-compatible APIs are also under development. Vertex Shader 1.1 and Pixel Shader 1.4 features are currently supported, along with the majority of features used by most developers when producing 3D games and applications.
SwiftShader can perform over 50 times faster than Microsoft's Direct3D Reference Rasterizer in tests with sample applications, and can rival the performance of low end hardware 3D graphics solutions in some cases. SwiftShader achieves this unprecedented level of performance by dynamically compiling highly optimized code specific to an applications 3D rendering needs.
SwiftShader is currently available for x86 CPUs with Intel's SSE multimedia instruction set extensions. SwiftShader runs on Microsoft Windows 98 and higher, and on Linux through TransGaming's Cedega portability technology.
OpenGL is an API, which programmers use to describe a set of graphics primatives.
SwiftShader is a renderer, which draws things.
You would, in fact, program your code in Direct3D or OpenGL, and then use SwiftShader as the renderer, the same way today you would program in Direct3D or OpenGL, and then use your ATI X800 as the renderer. They even mention, in the article, that "OpenGL-compatible APIs are also under development".
The only difference is that, compared to an ATI X800, SwiftShader will be very slow, and compared to the SuperImageCrazyMagic 9000 VGA+++ graphics card in my crappy laptop, SwiftShader will be quite fast.
Obviously I realise that a lot more is needed before desktop Linux taxes off, but if someone could capitalise on this we could have a decent GUI utilised without pissing all over Linux's reputation for not taxing hardware too heavily. (Personally I prefer an understated GUI which uses no resources, but obviously there is a market for eye candy.)
My 3D Texturing Skinning work (under construction)
So would something like this make it possible for me to play my MAME rom of 'San Francisco Rush' (jeah)? I cannot use my video card to play it and it runs like crap off software currently.
I realize that I'm just a humble AC, but let me say this: I'm not certain I understand the point here. Casual gamers are exactly that, casual. They're more interested in online play like Java Applets or MIDP Applications for phones than they are in Gee Whiz Bang games that they must install to their computers. So from that perspective, I really don't understand what market TransGaming plans to attack with this new software.
Even if we assume that casual gamers are looking to install games onto their computers, it's hard to say that such gamers wouldn't have at least a basic 3D card to play games on. Even the Intel Integrated Graphics (about as bad as you can get) has decent 3D support. Are the Gee Whiz features of DX9 really all that important to these players?
Formerly known as slow rendering.
http://www.transgaming.com.nyud.net:8090/swiftshad er.php
With no free licensing for personal and/or academic use available, I don't see that happening. Furthermore, it's only supported on Win 98 and up and Linux. One of the principal strengths of OpenGL is that it works on all Windows flavors, Linux, Unix, Solaris, Mac, ... -- Paul
OpenSource.MathCancer.org: open source comp bio
...saying that it performs up to 50 times better than the Microsoft Reference Rasterizer you're actually calling it slooooow. The REF driver exists with absolutely ZERO optimizations explicitly for the purpose of discerning if a problem is in your code or is in the video card's device driver. Maybe they're confused about the old RAMP and RGB devices you could use to render in software. 50 times faster than REF is pathetic to be honest.
Loading...
Just remember that there are no miracles. If it's ultra-fast
:)
and all, then it is either in fact a kick-ass uber-optimized
code, which is totally cool, or it is fast in _some_ cases
and it's still needs 'improvements'.
I may sound like an ass, but there ARE no miracles. Not in
computer graphics at least. Not since Bresenham and Doom
Please prove me wrong.
...an uber ultra-fast fixed-point math library!
Stating on Slashdot that I like cheese since 1997.
I'm sure that creating this engine was an interesting challenge .. but .. why? 2D only graphics cards don't really exist anymore. It doesn't exist for non-PC platforms so it doesn't really aid portability (though they say in the FAQs it could if someone wrote a "SwiftASM" thing for the target CPU) either.
Is it just a fun toy? Or have I missed something?
http://twitter.com/onion2k
Sure, it *may* be faster than other software rendering solutions, but it still only approaches (that's such a broad term) the performance of low-end cards: "can rival the performance of low end hardware 3D graphics solutions in some cases." Sounds pretty iffy to me. It'll be good in some cases; however, as it is, I doubt it will replace dedicated hardware cards for many people (it's not like low-end cards are very expensive. You can find decent (albeit older) graphics cards for under $20.
I'm not saying this technology isn't useful, it just has limited application in its current state.
This looks like it is meant to compete with Pixomatic from RAD Game Tools. ( http://www.radgametools.com/default.htm ) Perhaps it's cheaper or faster, but pixomatic is not overly pricy and I trust Mike Abrash *now at RAD) has a little bit of experience writing fast renderers :)
-- Too lazy to get a lower UID.
With no free licensing for personal and/or academic use available, I don't see that happening.
It's not intended for personal and/or academic use. It's yet another implementation of certain standards and presumably it works better than others on _certain_ platforms. Get it ?
Can it split the rendering load between your GPU and your CPU if your GPU is capable of some of these features? I couldn't find an answer on their website.
What about Wu's line-drawing algorithm?
Laws do not persuade just because they threaten. --Seneca
Believe it or not, but integrated graphics hold the lion's share of the PC graphics market. Nvidia and Ati are both pretty far behind Intel in terms of install base. This could be very bad for the other vendors: the main reason for the popularity of integrated graphics is cost - Intel itself only holds about a $5 premium on gfx-enabled chipsets over discrete chipsets.
What happens when Microsoft licenses this tech and integrates it into Windows? Suddenly, all anyone needs is a RAMDAC to output framebuffer to VGA, so Intel doesn't need to develop GPUs anymore, and overnight gets a massive performance boost and full DX9 support....
------- "From bored to fanboy in 3.8 asian girls" ----------
The latest Enlightenment is suppose to have a fast "software only" mode. I guess no one sees the point their either.
This will allow mixed redering, so that the old 3d card in a given system can do most of the work, and this will just render the things that require modern features. That would be quite useful.
The toad can't burp - and for some reason can't fart either, so it swells up and eventually explodes. --Anonymous Coward
Earlier this afternoon, the CEO of TransGaming was found on his office floor with two broken knees. Witnesses observed a pickup truck with the word "nVidia" printed on the side leaving the scene.
I, for one, welcome our new karma-whore sig writing overlords
You could accelerate 3D rendering, but using hardware! I know it sounds crazy, but it just might work. Who do I call to get my idea patented?
Anyone have time to, say, download this driver and fire up UT2004 or somesuch and test the framerate using software rendering vs their 3d card (with all other settings being equal)? Of course, this wouldn't be particularly scientific, but it would at least give some idea about how well this thing performs and whether it's useful.
IS this what they have been doing with my MONEY for Cedega???? Instead of releasing updates?? There hasnt been a cedega update (besides for a guild wars hotfix) in fricking 2 months!!
that was awesome
If this is an optimization, is there any chance of using some of the Swiftshader code to optimize hardware renderers?
No. Not unless you have an uber processor capable of running your game code and all the 3D bits at the same time with decent performance. Current systems when using software rendering are considerably slower than even a crap integrated graphics chipset. Thats not even with any shaders so don't think this would speed anything up.
What this technology could help with is for people with older 3D accelerator cards which don't support the latest whizbang features or for usage under Linux/whatever where you don't know how to access the accelerator features due to lack of docs. Performance under both cases wouldn't be amazing but it'd be better than nothing.
...is what demogroup they were before they went commercial, because that's the only crowd I can see with the drive and desire to create something like this. One of my group's coders still gets a stiffy for software rendering, and I know he's not the only demoscener that does.
The site is /.'d, so I can't rtfa.
Swiftshader is based on the sw-shader project, which produced very good quality output very quickly using SoftWire to compile the rasterizers. A lot of other software 3D implementations only optimize the most common cases and fall back on very slow, general purpose rasterizers to do the rest, often using giant switch statements or function pointers in their innermost loops to handle the countless combinations of blending, lighting, and raster options available. Even precompiling them all with a generator script or clever use of macros is infeasable due to the number of combinations, and just one of those will slow any 3D rendering to a crawl, which is the problem that sw-shader solved, by optimizing all cases.
What's good is that the project is once again under active development, and it's no longer windows-only. The downside is that it's gone commercial. With so few contributors other than the original author, that sort of thing can happen to an OSS project. He put a lot of hard work into it though, writing a substantially complete DirectX 9 replacement based on his library. Transgaming actually had to purchase two projects for this, because sw-shader depends on SoftWire.
SwiftShader code seems to be directly based on SoftWire and swShader, which used to be both SourceForge projects.
--
Open source drivers like nv (NVidia) are 2D only. With this toy you get at least some 3D-functions with such drivers.
With my new Quad Xeon 3.6 GHz, this enables me to play solitaire at a staggering 30fps, with only minor drops! Great! ;)
on fps in WoW or CS:S while in linux land?
lose != loose
What you're suggesting is that Microsoft might license tech from a company whose main product is a re-implemenation of the Microsoft Windows SDK for *NIX. If this happens, I'll either eat my hat or die of laughter. Oh the irony.
Games that require advanced shading tend to be CPU bound. If the CPU is your bottleneck, what good is it to give the CPU more work?
Like clickthrough rates on Slashdot articles are any higher.
Justice is the sheep getting arrested while an impartial judge declares the vote void.
Claiming that it's 50 times faster than some other software renderer is like claiming your racecar is 50 times faster than any other racecar that has no wheels.
What's the point?
I've been wanting to write an article or blog about that for a while now. With DVI digital you don't even need a DAC, just a bit of hardware to transfer data from RAM to the DVI connector with the right timing. Even better for Intel/AMD is that this provides an excellent use for dual cores. Most software won't use a second core, but you could use it as a GPU substitute by writing something like this. Dual cores for lower cost systems ;-)
Why are you thinking they're CPU-bound? All the shading happens on the graphics card. In fact if you look at the minimum system requirements, most games today require at most 1.5 GHz. Gameplay and AI have changed very little since 5 years ago. If you don't beleive me try playing any game and lowering the resolution and the graphics settings to the minimum, then note the increase in FPS - why would FPS increase so drastically if the game was CPU-bound?
Isn't this somewhat like how directx started out? Now if this thing can automatically off-load any processing on the built-in GPU and still keep up its performance, then this could be the beginning of cross platform 3D gaming which would mean the beginning of a widespread move to Linux.
I think we should christen this the Decade of Linux. Things have come very far from where they were in 2000 - Linux is now being shipped on OEM machines. I think its safe to say that by 2010, Linux gaming will be a reality and will be on par with Windows, if not better.
Add that to things like distros standardizing and striving for compatibility, usability studies being conducted that help in improving the end user's Linux experience and the availability of great development platforms like Eclipse, and you have the components in place.
So I think these 10 years will mark Linux's rise and will stabilize Linux as a widely accepted operating system and an alternative to Windows. I couldn't be bothered whether it marks Microsoft's fall or not, but I'm sure we will see more widespread Linux adoption.
As always, never forget the people who started this in the first place. They are the giants on whose shoulders Linux now stands.
Find a job you like and you will never work a day in your life.
Isn't this tech slower than integrated graphics? It seems that way to me. It's faster than the DX9 reference rasterizer, but that is called a reference rasterizer for a reason. Integrated graphics are designed to be pretty fast.
If a Linux GUI (ho ho ho) can provide an experience as rich as Aqua or Aero and base it on this software rendering
Something similar already exists. It's called Enlightenment (DR17).
Watch the video for a quick demo. There are also decent screenshots, though like with OS X, really don't do it justice.
Disclaimer: I'm both a OS X and Enlightment user (former KDE).
"It ain't a war against drugs.it's a war against personal freedom" --Bill Hicks
(Do we REALLY need aliens with realistic fur that whips in the wind as they make their slow approach toward the ground?)
If high-end graphics can be done entirely in software on a reasonably current machine WITHOUT having to spend hundreds on a separate board, software sales increase significantly.
Personally, I use a notebook computer - not expandable in the graphics department. Ultraportability has priority over graphics for me. I _would_ like to play recent games, but can't because they're made specifically for XYZ UltraCoolGraphicsSuperCard - which happens to be bigger than my computer. As such, I'm stuck playing Quake III, Oni, and Daikatana - would that these recent games had a kick-butt software renderer that let them run decently on a no-hardware-acceleration graphics card.
And that's where TransGaming comes in: I am a casual gamer looking to install games onto my computer - and I don't/can't have what is currently considered a "basic 3D card" to play games on. The better the game looks, the more likely I am to buy it (hey, you have your priorities, I have mine) - IF there is a software renderer available to deliver >90% of the Gee Whiz features of a DX9-compliant card, without the card.
What casual gamers have is a computer of generating images in software and painting them on the screen at 30FPS. Beyond that, the more common 3D hardware available (i.e.: Intel Integrated Graphics) is just too inconsistent and obsolete to bother supporting fully.
I don't want a 3D card. I have no place to put it on my Sony T-series ultraportable. Give me a kick-butt software renderer and a compatable game. TransGaming is providing the renderer; what are you providing? I'll give TransGaming money for a good renderer coupled with a recent game; I won't give you money for saying "so put up with cheesy graphics already".
Can we get a "-1 Wrong" moderation option?
Pull the other one, it's got bells on. You expect us to believe that the slashdot editors actually edited a submission?
- None can love freedom heartily, but good men; the rest love not freedom, but license. -- John Milton
This has been done at least twice before. NVidia, when they release new drivers, implement new features in software for older hardware (at least they have done so in the past).
Mesa 3D http://www.mesa3d.org/ has supported vertex and fragment programs since 2003.
Can it take advantage of multiple processors?
For years, some analysts claimed that ordinary processors would eventually obsolete 3D accelerators, because they would be fast enough to handle all of the rendering in software. Since graphics processing can usually make pretty good use of parallelism, then perhaps a package like this along with multiple CPUs is the "wave of the future"?
Obviously not now... but in 20 years?
I, for one, welcome our pixel shading overlords.
On the other hand, in Soviet Russia, the pixel is shadowing you.
Expect a beowolf cluster of SwiftShaders to run the matrix soon.
I'm still trying to figure out what people mean by 'social skills' here.
...to shell out $60-$100 for a cheap DX9 video card when you are paying ~$0 per new game. :P
You sound like they've never done anything stupid before...
Write boring code, not shiny code!
Any sugguestions as to actual game demos that will work with this? The little render demos that come with this are nice and all, but its preformance with a real world game would be nice.
"I'm not high, just stupid" --JY
As of 5:30 EST transgaming.com gives no reply to pings and webpage requests.
Regarding the lowend non-integrated gpu death: they are already dead. There's no chance of being able to compete with Intel in this area, just because "integrated" means much cheaper production. Shared VRAM, no PCB, even no need for a separate GPU chip! It costs nothing, it sells for nothing, and it works fine. If you don't want modern games or some other complex 3D software, you will be completely satisfied with it.
Ok, I read the page and the commentary already posted on /. and I'm still confused.
/. crowd, why is this ad for a niche commercial product that only runs on Windows (runs under Cedega doesn't count as it is also a commercial product) on slashdot as an article instead of a paid advertisement?
1. Who is the target market for this product? It is a software renderer. Might be faster than Mesa but is still going to run like a dog for any real world use. After all, just how many people have flaming P4s or A64s and a 2D only framebuffer? I suspect a Radeon 9200 with the 3d support in x.org would outperform this product.
2. Since the answer to the first question is almost certainly going to describe a fairly small group, even among the
Democrat delenda est
Suppose some number of years down the road, a lot of low-end mass market machines are using multicore processors (so the extra CPU load will be on another processor and won't throttle the game's performance too much). Would you get any kind of real speed boost from the fact that you could draw your textures and data directly from main memory, instead of having to push it over the PCI bus and cache it in graphics memory? Or will the busses be fast enough and memory cheap enough at that point that it won't make a difference? Or does it all depend too much on how this nonsense is coded?
Obviously it wouldn't make enough of a performance difference that people will start ditching their graphics cards; I'm just curious if it really buys them anything on texture-intensive games.
~fin
Where are mod points when you need them. That was pretty insightful.
I wonder how many frames/hour I can get!?
Link please?
I remember when legal used to mean lawful, now it means some kind of loophole. - Leo Kessler
Transgaming is due to release Cedega 5.0 for Linux in about 19 days.
With SwiftShader supported in Cedega, Transgaming can essentially claim to support the 'perfect' play of any DirectX 9 and below game from the Windows platform on Linux. This is a very huge step and fulfills rumors that the new Cedega 5.0 would have exactly this... 'playable support for *all* Windows games'. Now the problem is... it just won't run that fast. But it PLAYS.
I used to be a Cedega subscriber. I was creating my own builds from CVS for a while using a poached Gentoo ebuild, and eventually signed up so that I could vote and try to swing things away from that "must port Counter-Strike and EverQuest and nothing else" mindset that the Cedega community have.
Graphically, the thing is great. Games like Diablo 2 ran remarkably smoothly in most cases. But the game Oni, for instance, which was one of the more simple games that you would expect to run, won't run under Cedega with sound turned on, thanks to Cedega's shit support for DirectSound3D.
However, the way it works in the Cedega community, is that votes go on the games, not on the bugs. Never mind how many other games are affected by this one bug, if they're all separate games, each of which a few people like, the votes get spread out and games like EverQuest get all the attention. Hell, I saw them putting attention on Doom 3, when there was a week (A WEEK!) left until the native Linux release.
Reasons like this were what made me stop paying for the subscription... I'm sure they'll figure out the Right Way of handling voting at some point in the future (please, take some cues from Bugzilla and basically any other open source bug tracking, okay?) but until then, I'll be content with dual booting to play my Windows games even though it means turning off my torrents..
Anyway, it's comforting to see that they're still focusing entirely on the graphics end, while the sound subsystem silently rots away.
Karma: It's all a bunch of tree-huggin' hippy crap!
I wonder if this is possibly a take off of an open source project called swShader (its on source forge.) I've had some contact with the author of swShader, Nicholas Cappens(spelling likely incorrect), we discussed the unique approach he has taken to polygon filling/texturing. swShader also has DX 8 & 9 interfaces and a dll which could stand in to interface with regular DirectX games. While its an interesting piece of work, I hate to see when an OSS project, particularly one with one primary author(the owner if you will) gets taken and then commercialized becuae I'm sure he'll never see a penny, despite having written most of it. I don't know that thats the case here or not, but I hope it isn't. I guess its part of the deal when you open it up to OSS but it would still stink. Then again, maybe he's commercialized it himself or got hired to produce it, which I would be all for.
I'm confused. . . I thought TransGaming only made stuff for emulating Windows games on other systems. But it looks from the preview like this is a Windows program. . . why would someone who already can run Windows games want better software acceleration? Any card made in the past 4-5 years costing more than $70 or so supports hardware acceleration, and anyone with a computer older than 4-5 years old probably can't play games that require vertex and pixel shading, anyways.
www.linuxpenguin.net
You should start laughing while you're eating. Just in case.
I feel fantastic, and I'm still alive.
You have to wonder though, why the runtime code generation? That takes cycles to figure out what to do; and then if the environment changes, the code has to be rebuilt again. Run-time code generation looks like this to me:
- Decide what to do
- Build the pipeline
- Execute the pipeline
- Decide if the pipeline is still valid
- If true, execute the pipeline
- If false, go back to 0
Without it it looks like this:
- Decide what to do
- Execute those decisions
- Go back to 0
Also, runtime code generation is a bad solution; it means you have to be open to runtime code injection attacks if any exist that don't require advanced knowledge of address space. This isn't typically a problem, except now you have a program that expects known addresses to be code; a slight bump might give you a way to say "write X into your generated code, wherever that is."
I'm going to stitch together a small proof-of-concept, a make-believe "bytecode interpreter" that uses a different method of "pipelining" not requiring RCE. Let's see if I can find another solution.
Support my political activism on Patreon.
Vertex shaders, sure. In fact, DirectX does just that. If you lack hardware vertex shaders, DirectX will happily do them in software. It's not as fast, of course, since it hits your CPU, but it works. However other things can't be done, at least not yeat. Pixel shaders would be an example. They are later in the chain, after the graphics card has done it's work, even if it's a simple one. So that either requires doing it all in software, or sending data from the graphics card back to the CPU, doing more work, then back to the graphics card.
With AGP it's right out. PCIe makes it a remote possibility, the bus would work, but it's still not a real solution. It would end up being too slow to matter. The thing is you can get a cheap 3d card that WILL do it, though slow (any Radeon 9200 or above), but faster than software.
So I don't know about killer app. Certianly not for Windows. As it stands, DirectX does this to the extent that it's useful. Things that can be done in software to a reasonable speed will be, if necessary. Things that can't are simply excluded. It's probably the best compramise from a consumer standpoint.
...when pci was vogue, and the pentium was king(in X86 land) video cards had VRAM, and you were lucky to have more than 1MEG. After all 1024x768x24bit is about ~3.8 meg so these werent even capable of full SVGA at 24bit. But if you were lucky they worked ok at 640x480@24bit = ~921k ... ;) Even a 386 supported linear frame buffers in protected mode. This was used in games such as Zone66, which was incredibly fast, and required you to boot into the game, where it ran exclusively in protected mode, an optimization not available today under XP, where we have "update available" or "disk is low on space" popping up constantly even when playing a full screen game like doom3.
anyway, my point (sorry). IIRC the whole point of VRam was that it was dual ported, ie the arrangement of gates/junctions that make up the memory cell allowed the memory to be accessed both by the PCI bus or DMA controller, and the DAC's on the video card _at_ _the_ _same_ _time_ !11
It would be interesting to understand how the current state of the art compares. I believe DDR3 wraps the clock wire around twice to double the clock speed or some trick (remember all those RamBus patient battles?). Have we come in a cycle , like with parallel/serial buses used in disk storage, with RAM not having architectural embelishments in the latest itterations but rather running faster due to a more basic stripped down design? Or does sophistication increase monotonically? Nah, dont think so d:D
Not all systems come with good hardware rendering. And even though an upgrade is relatively cheap, not everyone is willing to pay that price for casual gaming, tons of people don't know/care how to do a graphics card upgrade, and many systems simply can't be upgraded (laptops in particular). TransGaming is all about portability. Whether from Windows to Linux, Mac to PlayStation, hardware to software... I can hardly imagine a better company for releasing a product like SwiftShader.
So here's something I don't get: Both projects used to be LGPLed, but now there are no source files for those old versions anywhere... does this mean a project can effectively be un-open sourced?
Someone should mirror these files! I just checked the license included in a few tar files and zip files, and it is the LGPL (in accordance with what is stated on the project pages). So anybody is allowed to copy and redistribute them.
It's a pity that TransGaming has decided to remove these files. If someone has the resources to mirror them (I don't) or to re-open the SourceForge project pages, then this should be done as soon as possible!
Crystal Pepsi was pretty cool.. although to my taste buds, it tasted a little more like Diet than regular.. Maybe they had to resort different means of flavoring that wouldn't color the syrup? *shrug*
http://undecidedgames.blogspot.com
This is one thing I'd like to see more of on Linux. It has lots of cool device drivers for things like webcams and firewire cards and advanced audio cards, joysticks, etc. But if you want to write a program that supports all those things, then you generally have to spend money on the hardware. If we want lots of developers to be able to contribute, and to develop apps as hardware becomes consumer-priced, then emulation would be a real step in the right direction.
I think I'd rather get a Geforce2 for $20 than pay $5 a month (I believe that's what the Cedega subscription costs) - and I doubt that a casual gamer is going to be willing to pay more for software emulation than a hardware upgrade. Upgrading your video card isn't hard - check the system's specs when you're buying the card to make sure it'll work, then pull the old one out and put the new one in.
And some laptops do allow you to upgrade the graphics card (from what I've heard, you can do this with many Dells - although these graphics cards are hard to find in stores but you might have luck on eBay or something).
www.linuxpenguin.net
While the casual gamer might not care much about shader effects itself, if the application uses it and his hardware doesn't support it he can't run it. That he does care about. And nowadays even simple kid's games start to use shaders, just because it's easier to develop. A Voodoo 3 does not support pixel shaders, neither does a Geforce 2.
SwiftShader won't need a subsription as far as I know. Game developers buy a license for a particular game and distribute a copy of SwiftShader with it. The people buying the game probably won't see any difference in price, or even a lower price, because SwiftShader is intended to reduce support costs for game companies. Anyway I don't know the details. If you're interested please constact sales@transgaming.com.