You would say that if you want to use traditional CAD methods like radius/chamfer instead of extrude/bevel etc.
You would definitely say it if you wanted to create manufacturable models and communicate their parameters to a manual or automated machining process.
You would say that if you need parametric modelling based on aspects of the model itself - e.g. you want your modelling tool to retain the relationships between entities in your model automatically - e.g. a dowel-piece is always.005mm smaller in radius than the hole it fits into etc. etc.
Trimmed NURBS are about the only primitives Maya/Houdini share with 'real' solid modelling tools, and their implementation is focussed on a completely different use-case.
CSG on parametric solids is a different kettle of fish from mesh-based CSG - and it is way, way more restrictive when it comes to organic modelling and fast workflow - which is what apps like Maya and Houdini are all about.
Mesh-based modellers (and most NURBS-based modellers that are useful for general purpose modelling) do *not* cater to CAD-like construction and output (e.g. dimensioned drawings, constraint-based modelling) workflow, least of all apps like Maya. Houdini has a much more procedural approach, but i don't see anybody doing large-scale rapid protyping, mechanical design or engineering analysis with Houdini. If you'd like to provide an example of an organisation that switched from AutoCAD, Pro/Engineer or SolidWorks to Houdini/Maya for mechanical design purposes - go right ahead.
Parametric solids-based modelling for mechanical design uses a completely different approach than surface-based modelling for animation/presentation, and there is a lot less of a crossover between the toolsets than you wqould think.
I use an XBox with a USB wireless LAN adapter for this purpose.
It runs Freevo, and while not completely silent, could be made so with a bigger CPU heatsink and a silent PSU- the noise it does make simply isn't intrustive enough for me to bother, however
I use Xebian, so no mods were required for the hardware (I did need to rend MechAssault to get Linux on the hard drive), and I can play XBox games if I want to.
Movies and MP3s are played off a samba share on another linux PC, and it mostly works pretty well.
There are a couple of caveats though-
Limux's approach to swapping memory is pretty awful for this type of application. After a long period of idle time, attempting to play a movie results in about a minute of vigorous disk-thrashing before anything actually happens on-screen.
The XPad driver/XBox controller seems to behave erratically - often requiring an unplug-replug cycle upon bootup to be recognised. Currently i dont have the XBox remote, but it can be used in place of the controller.
Disks sometimes get 'stuck' in the drive and won't eject - no matter whether the software 'eject' command is used or the hardware eject button.
Depending on how much time you have, all these issues could likely be fixed, but if i was you, i'd just buy a wifi set top box like this one:
http://www.dlink.com/products/?pid=318
The XBox is quite flexible, and it has been a fun toy to have, and does extra things like rip DVDs and act as a 'standard' Linux PC - but as far as a 'plug n play' solution, i doubt you will get what you want with any PC-based solution, at least without spending a lot of time tweaking - there are just too many things to go wrong, from boot time to peripherals to media error handling etc.
If it really bothers you that much, get the source, change the name of the app, and tell your friends about your cool new image editor.
They will, of course, all see the clear benefits of your approach over the existing development regime at gimp.org, and switch to using it and hacking on your tree. Word will get out and the masses will similarly adopt your project.
Sodipodi-> Inkscape is a similar transition, as is Xfree86 -> X.org.
If it really bothers you, step up and make the change, otherwise, youre just whining at the whiners.
He has acquired copies of hundreds/thousands of pieces of software, is using them for commercial gain, and doesn't want to pay for them for some reason that simply escapes me.
He needs to either negotiate with the software authors/publishers to get 'special dispensation' to use their copyrighted works in his specific application free-of-charge, or he needs to purchase the software outright.
Every creative work is subject to copyright - there really isn't that much of a grey area here.
People like SCO have recently confused the issues by outright lying and fraudulently representing their interest in software products, however this is a problem with 'due process', not with copyright.
What he really needs to be suing for is a clear definition of what 'Fair Use' is, and if his application falls outside that definition, then tough cookies, he's breaking the law and is liable for hundreds of thousands of dollars in dmages or jailtime, just like every other mom, pop and grandma with a pirated copy of Windows and Office on their machine.
Perhaps he should also be suing for higher evidentiary standards in copyright cases - certainly the abolition of copyright for computer software is totally inappropriate and would disenfranchise the vast majority of software rightsholders (including GPL and other Free/Open licensors)
All the professional 3D apps e.g. Maya, Shake etc. are built to operate in this kind of 'decoupled rendering' environment, and render farms are probably the simplest type of 'distributed computing' - generally theyre just a bunch of independent computers on a LAN which see a shared disk on which the source materials (e.g. textures, models, motion data, imagery) is kept.
Accounting is pretty simple too as a central dispatcher keeps track of which frame is submitted to which node, and how long it takes - it is uncommon to have 1 machine working on more than 1 frame simultaneously.
Sun also offer this type of service, as well as others. I think it will be interesting to see if this model is adopted by hollywood etc. but I can see a lot of smaller shops taking advantage of it to acheive what would otherwise be impossible.
I remember using a Compaq Test-Drive-Programme 4-CPU Alpha 21264 box to render a 4096x4096 cornell box using BMRT that would have taken my 128MB P3-500 an age to do.
The model definitely works from my point of view, however I don't really think that Sun or HP will be in much of a position to actually make money out of this type of service since hardware depreciates so fast and the price people will be willing to pay would be pretty low I imagine - I guess they could simply use their unsold inventories to extract some kind of revenue from them instead of writing them off, but getting the 'volume' up to profitable levels will be a challenge.
One of the iPods killer features is it's tight integration with iTunes.
Chenges you make on your PC are seamlessly carried over to your iPod when you sync it.
Podcasting etc. is something I have been pleasantly surprised at, and none of this is very straightforward with a CD-based player.
However, I do have a CD-based MP3 player in my car, and before I bought the iPod I had a portable CD-based unit.
Battery life on CD-based units is pretty good, and I would probably choose a CD-based unit while travelling etc, since it can take AAs and isn't such a target for theives in unfamiliar places.
If you have a significant CD collection, or havent got around to ripping it yet, a CD-capable player also makes sense.
Whether Apple will produce such a player, I can't say, but I suppose an interesting concept might be an addon for iTunes to allow you to 'sync' some playlists with the content on a multisession CD-R/CD-RW for use in a MP3-CD player.
Buy an SGI Octane2 with the uncompressed HDTV option and youre good to go. This is a 'Personal Computer' which will do exactly what you ask.
Nothing you can buy 'off the shelf at the mall' will do this however.
The PCI bus is simply inadequate for this task, and even PCI Express etc. is probably inadequate too.
Hopefully this kind of application will drive computer manufacturers to introduce faster interconnects, since the CPUs and memory busses are certainly fast enough these days.
Apple at least realises the value of solid engineering and a good user experience.
The iPod, while not jam-packed with features, is fast and stable.
Creatives products (at least the older ones i have seen) are slow and buggy.
The iPod is sleek and minimalist, Creatives products are covered with chrome trim and raised, plastic buttons with a little hole you have to push a paperclip into to reset it.
The iPod's elegance and simplicity extends to its custom written and polished software package. Creative just bundles whatever crap it can license the cheapest.
I gladly bought an iPod, I wouldnt touch a Creative player with a ten foot pole.
FC3 is good once you get it to go. It did render my system unbootable when first installed, however.
I boot of a SATA disk, and was running a FC1 system with a 'roll-your-own' 2.6.5 kernel.
When i upgraded to FC3, it seems that the SATA drivers are compiled as modules and aren't loaded at the correct time to make the system see the SATA drives correctly - thus the system would simply refuse to boot.
Booting on my old 2.6.5 kernel got the system up, and recompiling the redhat kernel from SRPM with the SATA driver compiled in fixed all this, but it seems a bit of an oversight to ship a distro that isn't bootable from SATA these days.
GStreamer did not install correctly, and FC3 is pretty crap out-of-the-box for playing sound/video etc.
Xorg produced strange noise and snow on my screen with the 'nv' driver, something i had never seen before on any distro.
There is a real problem configuring yum/up2date repositories - newbies shouldnt have to hand-hack files, and almost everyone will need to configure livna/freshrpma/atrpms etc. to have a decent system.
I also had to hand-hack xorg.conf to get NVidia Twinview working - not surprising but no newbie will be able to see how to do this.
On the plus side, FC3 seems much faster than FC1, and GNOME 2.8 is getting pretty good - Once installed, it seems yo 'just go', and stay mostly out of my way, which is how a desktop environment should work. Every other aspect of the system seems solid, and as someone who has been using RedHat since 5.x, it all seems very familiar and comfortable.
I would recommend it as a corporate desktop where support is there, or for a power user.
Well, when you got the X-Box for free (friend went overseas and gave it to me), yeah it's cost beneficial.
Plus, this is just a hobby for me, and while I agree with you that the X-Box is no speed demon for rendering, and is memory limited, the point is it is a versatile little machine, which can be easily clustered and is capable fo doing 'real work'
Of course you can get more 'bang for buck', but if you are capped w/regard to absolute expenditure - e.g. if you only have $500, then the XBox stacks up pretty well in my view.
Find an SGI Octane with an SDI/HDTV I/O board and a large, fast disk array and you can probably do this in reasonably straightforward fashion.
The Octanes have massive bandwidth and don't suffer from the PCI bottleneck with their video I/O hardware modules.
Octanes are pretty cheap these days on EBay, the HDTV I/O hardware probably won't be, but you never know what kind of deals you can turn up.
Now the next question is why you would want to do this? Uncompressed HDTV will fill up over 200GB/hour, better to save the compressed MPEG-2 stream and decompress only when you want to view/transcode it.
Do people in the US actually believe there is some vast terrorist army out there that the US army is fighting? and who is poised to 'Invade the Homeland', and that there can be any meaningful protection from people hellbent on destroying shit if theyre willing to give their lives to do it?
Cos thats just plain batshit crazy. Surely youre not all so stupid as to believe that the 'Department of Homeland Security' is not there to protect Americans from anyone but each other?
>Nah. All that interface is being slowly deprecated. >The latest trend is to just support OpenGL so that >the graphics cards and drivers do not have to >support two interfaces.
And if OpenGL alone really worked for complex, pixel-perfect imagery, typography and video I/O, I guess that would be true.
However, we have a long way to go before the OpenGL API and consumer 3D hardware has the necessary polish to perform these tasks properly.
I agree in principle with the idea that various APIs can be run on top of OpenGL to take advantage of the speed of 3D hardware for some aspects of their operation - and that in the presence of said 3D hardware, this makes sense.
However I can't see how that invalidates the approach of making a video accelerator that uses a different approach to accelerate those same functions, especially if it can be done without the expense and complexity of a 3D pipeline that sits idle for most common desktop operations.
I mean, copper cable networking is being deprecated in favour of fiber optics and wireless, but theres still a huge amount of business in ethernet,DSL, phone and T1/T3 lines etc.
Make an X.org accelerator - There are a lot of people who dont care much for pushing polygons, but would love to have a fast, high quality grahics card that intergrates with X.org or XFree86 and works without hassle.
Support multi-head operation with robust Xinerama support, good colour calibration etc. and provide hardware acceleration for compositing, video4linux overlays, SDL hardware blitting, X primitives, Freetype font renderers, DirectFB acceleration - this card could form the heart of every low-cost or embedded linux system sold in existing or emerging markets round the world, and provide significantly better 2D desktop acceleration than ATI or NVidia, who seem to put 100% of their efforts into appeasing the Doom3 players.
Even if its not a match 3D-wise to a Geforce FX6800, it wouldn't be hard to do a better job of supporting Linux APIs than 90% of the manufacturers out there.
Darl has run this once innovative and successful company so far into the ground that they see 'the competition' as PJ/Groklaw.
So now it's not about IBM, it's not about UNIX, it's not about Linux, it's not about 'Intellectual Property'.
Now it's about a lone ex-paralegal who had the balls (and i mean that in the nicest way possible) to tell it like it is.
We can't have that, can we Darl? God forbid anyone actually accept a version of events that corresponds with legal and technical realities instead of simply believing whatever stupid lies you cooked up after another hard night on the Canopy crackpipe.
Whats next, are you going to come up with an alternate justice system because no court in the US will accept SCO's ridiculous legal 'arguments' either?
People *hate* what you and your company are doing, Mr. McBride.
It is wrong, and no amount of P.R. spin will change that. Shame on you.
One of Sun's key arguments in its 'Solaris and JDS are superior to Linux because...' campaign is that Sun's products are indemnified against IP problems, and we can expect to see a Microsoft patent used against Linux in the near future - in a FUD atttack to drive people towards Solaris as 'The only safe choice in x86 *NIX'.
Microsoft wants all UNIX users corralled up behind a single company so they can then simply drive that company into the ground, instead of having to play 'whack-a-mole' with Linux distributors.
Sun are taking advantage of this by profiting when they can, but they must realise this is a business strategy that is assuming 'eventual defeat' - Sun are clearly not able to cut a path as an independent technology company, and feel that becoming the 'New Apple to Microsoft' - e.g. expect a Microsoft Virtual PC port with a bundled XP/Longhorn Licence to Solaris x86 soon - is the best way to ensure survival in the short to medium term.
Clearly, they have no long-term strategy, unless it is to simply cede their server market to Windows NT and fade quietly into oblivion like SCO.
If the USA doesn't move beyond Windows on the desktop, there are a lot of other countries who will - Software patents as implemented by the US government, are overwhelmingly stupid, and even if every US Linux distributor faces massive taxes, you will have to deal with the fact that Linux is as prevalent, and as easily developed in Europe or India, or Japan, as it is in the US.
Linux is not going out like this, and this whole 'intellectual property lawsuit' business is just making me, as a programmer, computer user and an educated, open minded person, really angry.
If companies like Redhat are making money out of selling something they also give away for free, this is largely a result of the groundswell of dissatisfaction with the crap we have had to pay so much money for up till now -
Make a good product using ethical business practices and provide clear benefits to a supporting community of users, with contracts based on mutual trust, not meaningless stockmarket numbers and the threat of litigation, and people will be interested in buying it, using it and developing it.
When did 'The network is the computer' get replaced by 'The Microsoft(R) Network (c) Microsoft Corporation 2004 is the computer (pat. pending)'?
Mazda's rotary engine is well suited to the combustion of hydrogen, not least because it completely separates the intake, combustion and exhaust stages - with a piston engine there is a lot of potential for catastophic backfire, and high performance without any valve overlap (which would somewhat prevent this) is difficult to acheive.
The renesis (side-ported intake and exhaust - 'normal' rotaries have peripheral exhaust and often intake ports and intake/exhaust port overlap is employed to maximise performance at high revs, resulting in the characteristic 'brap-brap-brap' pulsing idle of a race or drag rotary engine and incredibly poor fuel economy at low revs) rotary engine doesnt suffer from this problem, allowing high-revs, aggressive induction and exhaust port profiles, along withthe light weight and excellent power-weight ratio rotaries inherently possess.
The current hybrid engine in the RX-8 only produces about 120hp when operating on hydrogen which isn't exactly stunning, but bear in mind that the original RX-7 produced less than this, while the last model to roll off the production line produced in excess of 280.
400+ HP is relatively easily acheiveable with proper porting, fueling and turbocharging of the 1.3 litre 13B engine on petrol, and with further development (or even tuning for hydrogen-only operation) it is not too far fetched to imagine the hydrogen-powered rotary performing on par or better than conventional fuels.
More info can be found:
http://rotarynews.com/?q=node/view/216
and a hydrogen--powered RX-8 looks like:
http://www.ultimatecarpage.com/frame.php?file=pi c. php&imagenum=1&carnum=1792
Absolutely - upon reading my previous comment, I perhaps suggested that only coders can make a difference, which is not really what I meant.
It's attitude, and willingness to participate in communities that make OSS what it is.
Filing bug reports, providing help and support for others etc. are all valuable, and shouldn't be underestimated.
No, what I cannot stand is this idea that has been pushed on the world at large by the commercial software industry that you, as an individual, are utterly powerless to make a meaningful change in the way your computer works.
It's just not true, and anyone who does care, and is prepared to make some kind of effort is able to help, and contribute meaningfully.
Its the people that claim theres no point at all in trying to contribute, that attempting to actually do anything past waiting to be spoonfed by your keepers, corporate, OSS or otherwise is 'unrealistic' that p*ss me off.
You can keep saying it's unrealistic to expect users to help fix problems with OSS software, but the fact is that only people who do put in the effort make any difference.
The only people who can effect changes are people who do code, who don't accept this defeatist version of 'reality'. If everyone simply accepted it was unrealistic to be able to personally contribute to anything, well, this world would be a much worse place.
What is 'realistic' to this guy is just not relevant to OSS development. Thats what makes OSS different, and special.
No, it's not 'realistic', but its happening, and it's happening regardless of how 'realistic' you think it is.
Slashdot Mirror
You would say that if you want to use traditional CAD methods like radius/chamfer instead of extrude/bevel etc.
.005mm smaller in radius than the hole it fits into etc. etc.
You would definitely say it if you wanted to create manufacturable models and communicate their parameters to a manual or automated machining process.
You would say that if you need parametric modelling based on aspects of the model itself - e.g. you want your modelling tool to retain the relationships between entities in your model automatically - e.g. a dowel-piece is always
Trimmed NURBS are about the only primitives Maya/Houdini share with 'real' solid modelling tools, and their implementation is focussed on a completely different use-case.
CSG on parametric solids is a different kettle of fish from mesh-based CSG - and it is way, way more restrictive when it comes to organic modelling and fast workflow - which is what apps like Maya and Houdini are all about.
Mesh-based modellers (and most NURBS-based modellers that are useful for general purpose modelling) do *not* cater to CAD-like construction and output (e.g. dimensioned drawings, constraint-based modelling) workflow, least of all apps like Maya. Houdini has a much more procedural approach, but i don't see anybody doing large-scale rapid protyping, mechanical design or engineering analysis with Houdini. If you'd like to provide an example of an organisation that switched from AutoCAD, Pro/Engineer or SolidWorks to Houdini/Maya for mechanical design purposes - go right ahead.
Parametric solids-based modelling for mechanical design uses a completely different approach than surface-based modelling for animation/presentation, and there is a lot less of a crossover between the toolsets than you wqould think.
Currently, my XBox runs a 2.4 kernel, but that is useful and intersting info. Thanks
I use an XBox with a USB wireless LAN adapter for this purpose.
It runs Freevo, and while not completely silent, could be made so with a bigger CPU heatsink and a silent PSU- the noise it does make simply isn't intrustive enough for me to bother, however
I use Xebian, so no mods were required for the hardware (I did need to rend MechAssault to get Linux on the hard drive), and I can play XBox games if I want to.
Movies and MP3s are played off a samba share on another linux PC, and it mostly works pretty well.
There are a couple of caveats though-
Limux's approach to swapping memory is pretty awful for this type of application. After a long period of idle time, attempting to play a movie results in about a minute of vigorous disk-thrashing before anything actually happens on-screen.
The XPad driver/XBox controller seems to behave erratically - often requiring an unplug-replug cycle upon bootup to be recognised. Currently i dont have the XBox remote, but it can be used in place of the controller.
Disks sometimes get 'stuck' in the drive and won't eject - no matter whether the software 'eject' command is used or the hardware eject button.
Depending on how much time you have, all these issues could likely be fixed, but if i was you, i'd just buy a wifi set top box like this one:
http://www.dlink.com/products/?pid=318
The XBox is quite flexible, and it has been a fun toy to have, and does extra things like rip DVDs and act as a 'standard' Linux PC - but as far as a 'plug n play' solution, i doubt you will get what you want with any PC-based solution, at least without spending a lot of time tweaking - there are just too many things to go wrong, from boot time to peripherals to media error handling etc.
If it really bothers you that much, get the source, change the name of the app, and tell your friends about your cool new image editor.
They will, of course, all see the clear benefits of your approach over the existing development regime at gimp.org, and switch to using it and hacking on your tree. Word will get out and the masses will similarly adopt your project.
Sodipodi-> Inkscape is a similar transition, as is Xfree86 -> X.org.
If it really bothers you, step up and make the change, otherwise, youre just whining at the whiners.
He has acquired copies of hundreds/thousands of pieces of software, is using them for commercial gain, and doesn't want to pay for them for some reason that simply escapes me.
He needs to either negotiate with the software authors/publishers to get 'special dispensation' to use their copyrighted works in his specific application free-of-charge, or he needs to purchase the software outright.
Every creative work is subject to copyright - there really isn't that much of a grey area here.
People like SCO have recently confused the issues by outright lying and fraudulently representing their interest in software products, however this is a problem with 'due process', not with copyright.
What he really needs to be suing for is a clear definition of what 'Fair Use' is, and if his application falls outside that definition, then tough cookies, he's breaking the law and is liable for hundreds of thousands of dollars in dmages or jailtime, just like every other mom, pop and grandma with a pirated copy of Windows and Office on their machine.
Perhaps he should also be suing for higher evidentiary standards in copyright cases - certainly the abolition of copyright for computer software is totally inappropriate and would disenfranchise the vast majority of software rightsholders (including GPL and other Free/Open licensors)
All the professional 3D apps e.g. Maya, Shake etc. are built to operate in this kind of 'decoupled rendering' environment, and render farms are probably the simplest type of 'distributed computing' - generally theyre just a bunch of independent computers on a LAN which see a shared disk on which the source materials (e.g. textures, models, motion data, imagery) is kept.
Accounting is pretty simple too as a central dispatcher keeps track of which frame is submitted to which node, and how long it takes - it is uncommon to have 1 machine working on more than 1 frame simultaneously.
Sun also offer this type of service, as well as others. I think it will be interesting to see if this model is adopted by hollywood etc. but I can see a lot of smaller shops taking advantage of it to acheive what would otherwise be impossible.
I remember using a Compaq Test-Drive-Programme 4-CPU Alpha 21264 box to render a 4096x4096 cornell box using BMRT that would have taken my 128MB P3-500 an age to do.
The model definitely works from my point of view, however I don't really think that Sun or HP will be in much of a position to actually make money out of this type of service since hardware depreciates so fast and the price people will be willing to pay would be pretty low I imagine - I guess they could simply use their unsold inventories to extract some kind of revenue from them instead of writing them off, but getting the 'volume' up to profitable levels will be a challenge.
One of the iPods killer features is it's tight integration with iTunes.
Chenges you make on your PC are seamlessly carried over to your iPod when you sync it.
Podcasting etc. is something I have been pleasantly surprised at, and none of this is very straightforward with a CD-based player.
However, I do have a CD-based MP3 player in my car, and before I bought the iPod I had a portable CD-based unit.
Battery life on CD-based units is pretty good, and I would probably choose a CD-based unit while travelling etc, since it can take AAs and isn't such a target for theives in unfamiliar places.
If you have a significant CD collection, or havent got around to ripping it yet, a CD-capable player also makes sense.
Whether Apple will produce such a player, I can't say, but I suppose an interesting concept might be an addon for iTunes to allow you to 'sync' some playlists with the content on a multisession CD-R/CD-RW for use in a MP3-CD player.
Buy an SGI Octane2 with the uncompressed HDTV option and youre good to go. This is a 'Personal Computer' which will do exactly what you ask.
Nothing you can buy 'off the shelf at the mall' will do this however.
The PCI bus is simply inadequate for this task, and even PCI Express etc. is probably inadequate too.
Hopefully this kind of application will drive computer manufacturers to introduce faster interconnects, since the CPUs and memory busses are certainly fast enough these days.
Apple at least realises the value of solid engineering and a good user experience.
The iPod, while not jam-packed with features, is fast and stable.
Creatives products (at least the older ones i have seen) are slow and buggy.
The iPod is sleek and minimalist, Creatives products are covered with chrome trim and raised, plastic buttons with a little hole you have to push a paperclip into to reset it.
The iPod's elegance and simplicity extends to its custom written and polished software package. Creative just bundles whatever crap it can license the cheapest.
I gladly bought an iPod, I wouldnt touch a Creative player with a ten foot pole.
FC3 is good once you get it to go. It did render my system unbootable when first installed, however.
I boot of a SATA disk, and was running a FC1 system with a 'roll-your-own' 2.6.5 kernel.
When i upgraded to FC3, it seems that the SATA drivers are compiled as modules and aren't loaded at the correct time to make the system see the SATA drives correctly - thus the system would simply refuse to boot.
Booting on my old 2.6.5 kernel got the system up, and recompiling the redhat kernel from SRPM with the SATA driver compiled in fixed all this, but it seems a bit of an oversight to ship a distro that isn't bootable from SATA these days.
GStreamer did not install correctly, and FC3 is pretty crap out-of-the-box for playing sound/video etc.
Xorg produced strange noise and snow on my screen with the 'nv' driver, something i had never seen before on any distro.
There is a real problem configuring yum/up2date repositories - newbies shouldnt have to hand-hack files, and almost everyone will need to configure livna/freshrpma/atrpms etc. to have a decent system.
I also had to hand-hack xorg.conf to get NVidia Twinview working - not surprising but no newbie will be able to see how to do this.
On the plus side, FC3 seems much faster than FC1, and GNOME 2.8 is getting pretty good - Once installed, it seems yo 'just go', and stay mostly out of my way, which is how a desktop environment should work. Every other aspect of the system seems solid, and as someone who has been using RedHat since 5.x, it all seems very familiar and comfortable.
I would recommend it as a corporate desktop where support is there, or for a power user.
Well, when you got the X-Box for free (friend went overseas and gave it to me), yeah it's cost beneficial.
Plus, this is just a hobby for me, and while I agree with you that the X-Box is no speed demon for rendering, and is memory limited, the point is it is a versatile little machine, which can be easily clustered and is capable fo doing 'real work'
Of course you can get more 'bang for buck', but if you are capped w/regard to absolute expenditure - e.g. if you only have $500, then the XBox stacks up pretty well in my view.
My Xbox renders frames along with an Athlon 2600+. and an Athlon 750. I'm planning to add more to my
/audio streaming 'set top box'.
network once they drop in price even further.
The XBox will happily support USB wireless LAN etc, and running Freevo makes a fairly slick video
It rips DVDs at 10-15 frames per second so I keep it busy crunching away on my DVD collection when it's not being used for other things.
It will also play XBox games, but I don't own any of those.
Its really quite a useful little machine to have around, although the lack of 3D drivers really limits it's true potential.
Find an SGI Octane with an SDI/HDTV I/O board and a large, fast disk array and you can probably do this in reasonably straightforward fashion.
The Octanes have massive bandwidth and don't suffer from the PCI bottleneck with their video I/O hardware modules.
Octanes are pretty cheap these days on EBay, the HDTV I/O hardware probably won't be, but you never know what kind of deals you can turn up.
Now the next question is why you would want to do this? Uncompressed HDTV will fill up over 200GB/hour, better to save the compressed MPEG-2 stream and decompress only when you want to view/transcode it.
Do people in the US actually believe there is some vast terrorist army out there that the US army is fighting? and who is poised to 'Invade the Homeland', and that there can be any meaningful protection from people hellbent on destroying shit if theyre willing to give their lives to do it?
Cos thats just plain batshit crazy. Surely youre not all so stupid as to believe that the 'Department of Homeland Security' is not there to protect Americans from anyone but each other?
and glDrawPixels is taking advantage of the 3D hardware how?
>Nah. All that interface is being slowly deprecated.
>The latest trend is to just support OpenGL so that
>the graphics cards and drivers do not have to
>support two interfaces.
And if OpenGL alone really worked for complex, pixel-perfect imagery, typography and video I/O, I guess that would be true.
However, we have a long way to go before the OpenGL API and consumer 3D hardware has the necessary polish to perform these tasks properly.
I agree in principle with the idea that various APIs can be run on top of OpenGL to take advantage of the speed of 3D hardware for some aspects of their operation - and that in the presence of said 3D hardware, this makes sense.
However I can't see how that invalidates the approach of making a video accelerator that uses a different approach to accelerate those same functions, especially if it can be done without the expense and complexity of a 3D pipeline that sits idle for most common desktop operations.
I mean, copper cable networking is being deprecated in favour of fiber optics and wireless, but theres still a huge amount of business in ethernet,DSL, phone and T1/T3 lines etc.
Make an X.org accelerator - There are a lot of people who dont care much for pushing polygons, but would love to have a fast, high quality grahics card that intergrates with X.org or XFree86 and works without hassle.
Support multi-head operation with robust Xinerama support, good colour calibration etc. and provide hardware acceleration for compositing, video4linux overlays, SDL hardware blitting, X primitives, Freetype font renderers, DirectFB acceleration - this card could form the heart of every low-cost or embedded linux system sold in existing or emerging markets round the world, and provide significantly better 2D desktop acceleration than ATI or NVidia, who seem to put 100% of their efforts into appeasing the Doom3 players.
Even if its not a match 3D-wise to a Geforce FX6800, it wouldn't be hard to do a better job of supporting Linux APIs than 90% of the manufacturers out there.
Darl has run this once innovative and successful company so far into the ground that they see 'the competition' as PJ/Groklaw.
So now it's not about IBM, it's not about UNIX, it's not about Linux, it's not about 'Intellectual Property'.
Now it's about a lone ex-paralegal who had the balls (and i mean that in the nicest way possible) to tell it like it is.
We can't have that, can we Darl? God forbid anyone actually accept a version of events that corresponds with legal and technical realities instead of simply believing whatever stupid lies you cooked up after another hard night on the Canopy crackpipe.
Whats next, are you going to come up with an alternate justice system because no court in the US will accept SCO's ridiculous legal 'arguments' either?
People *hate* what you and your company are doing, Mr. McBride.
It is wrong, and no amount of P.R. spin will change that. Shame on you.
One of Sun's key arguments in its 'Solaris and JDS are superior to Linux because...' campaign is that Sun's products are indemnified against IP problems, and we can expect to see a Microsoft patent used against Linux in the near future - in a FUD atttack to drive people towards Solaris as 'The only safe choice in x86 *NIX'.
Microsoft wants all UNIX users corralled up behind a single company so they can then simply drive that company into the ground, instead of having to play 'whack-a-mole' with Linux distributors.
Sun are taking advantage of this by profiting when they can, but they must realise this is a business strategy that is assuming 'eventual defeat' - Sun are clearly not able to cut a path as an independent technology company, and feel that becoming the 'New Apple to Microsoft' - e.g. expect a Microsoft Virtual PC port with a bundled XP/Longhorn Licence to Solaris x86 soon - is the best way to ensure survival in the short to medium term.
Clearly, they have no long-term strategy, unless it is to simply cede their server market to Windows NT and fade quietly into oblivion like SCO.
If the USA doesn't move beyond Windows on the desktop, there are a lot of other countries who will - Software patents as implemented by the US government, are overwhelmingly stupid, and even if every US Linux distributor faces massive taxes, you will have to deal with the fact that Linux is as prevalent, and as easily developed in Europe or India, or Japan, as it is in the US.
Linux is not going out like this, and this whole 'intellectual property lawsuit' business is just making me, as a programmer, computer user and an educated, open minded person, really angry.
If companies like Redhat are making money out of selling something they also give away for free, this is largely a result of the groundswell of dissatisfaction with the crap we have had to pay so much money for up till now -
Make a good product using ethical business practices and provide clear benefits to a supporting community of users, with contracts based on mutual trust, not meaningless stockmarket numbers and the threat of litigation, and people will be interested in buying it, using it and developing it.
When did 'The network is the computer' get replaced by 'The Microsoft(R) Network (c) Microsoft Corporation 2004 is the computer (pat. pending)'?
Mazda's rotary engine is well suited to the combustion of hydrogen, not least because it completely separates the intake, combustion and exhaust stages - with a piston engine there is a lot of potential for catastophic backfire, and high performance without any valve overlap (which would somewhat prevent this) is difficult to acheive.
i c. php&imagenum=1&carnum=1792
The renesis (side-ported intake and exhaust - 'normal' rotaries have peripheral exhaust and often intake ports and intake/exhaust port overlap is employed to maximise performance at high revs, resulting in the characteristic 'brap-brap-brap' pulsing idle of a race or drag rotary engine and incredibly poor fuel economy at low revs) rotary engine doesnt suffer from this problem, allowing high-revs, aggressive induction and exhaust port profiles, along withthe light weight and excellent power-weight ratio rotaries inherently possess.
The current hybrid engine in the RX-8 only produces about 120hp when operating on hydrogen which isn't exactly stunning, but bear in mind that the original RX-7 produced less than this, while the last model to roll off the production line produced in excess of 280.
400+ HP is relatively easily acheiveable with proper porting, fueling and turbocharging of the 1.3 litre 13B engine on petrol, and with further development (or even tuning for hydrogen-only operation) it is not too far fetched to imagine the hydrogen-powered rotary performing on par or better than conventional fuels.
More info can be found:
http://rotarynews.com/?q=node/view/216
and a hydrogen--powered RX-8 looks like:
http://www.ultimatecarpage.com/frame.php?file=p
Sun have about as much chance of impacting Linux's momentum as SCO do.
I mean, what are they gonna do, refuse to release the specs for their new CPUs so Linux can't run on them?
I bet the managment at Redhat are losing lots of sleep over that.
But this is no dream of mine.
I'm not going to get excited about another rubbish game made to cash-in on the success of the LOTR movies.
There have been at least 4 already, whats so special about number 5?
Absolutely - upon reading my previous comment, I perhaps suggested that only coders can make a difference, which is not really what I meant.
It's attitude, and willingness to participate in communities that make OSS what it is.
Filing bug reports, providing help and support for others etc. are all valuable, and shouldn't be underestimated.
No, what I cannot stand is this idea that has been pushed on the world at large by the commercial software industry that you, as an individual, are utterly powerless to make a meaningful change in the way your computer works.
It's just not true, and anyone who does care, and is prepared to make some kind of effort is able to help, and contribute meaningfully.
Its the people that claim theres no point at all in trying to contribute, that attempting to actually do anything past waiting to be spoonfed by your keepers, corporate, OSS or otherwise is 'unrealistic' that p*ss me off.
You can keep saying it's unrealistic to expect users to help fix problems with OSS software, but the fact is that only people who do put in the effort make any difference.
The only people who can effect changes are people who do code, who don't accept this defeatist version of 'reality'. If everyone simply accepted it was unrealistic to be able to personally contribute to anything, well, this world would be a much worse place.
What is 'realistic' to this guy is just not relevant to OSS development. Thats what makes OSS different, and special.
No, it's not 'realistic', but its happening, and it's happening regardless of how 'realistic' you think it is.
Really, what other legitimate use does a 30Mbps connection to the home have than the large-scale stealing of copyrighted material?