Speed is obtained by doing it for yourself down to the barest Win32 API layers. Power is obtained the same way. MFC buys one thing- laziness. It's easy to bang out a UI for an office/buisness app with it. Try to do something fancy with it- forget it.
WINE derived apps are slower than many native ones. Why? Because it's translating things over to native behaviors. It's still a translation layer, albeit a thin one.
Which would I rather have? If I had a choice of no app or a WINE app? Dumb question. If I had the choice of a native Qt, Fltk, Wx, JX, GTK+, etc. app versus a WINE app? Dumb question, again- native is better than non-native.
If it's multithreaded or multiprocess, SMP works.
on
Pentium 4 Delayed
·
· Score: 2
If the game's got multiple threads or has a seperate server daemon process, the SMP machine will edge out the uni-processor one if the SMP machine's CPUs are anything faster than half the uni-processor's CPU. Same goes for anything out there that has threads or has external processing applications. I don't want a 1GHz machine- I'm perfectly happy with a dual or quad 750 for now.
Greetings! I couldn't remember which big paper it was- Thanks!
Keep plugging away at these clowns- they deserve every drop of agony that's about to come to them.
That's partly because of the implementation...
on
3dfx Does OpenGL
·
· Score: 2
If you've got a RagePRO, it's merely a matter of adding a "Module" section with a 'Load "glx.so"' in it, copying/linking 4 files in the right places (Three files/links go to/usr/X11R6/lib- all the libGL* files. One file goes in/usr/X11R6/lib/modules- glx.so.) and you're good to go with Utah-GLX.
It is literally that simple.
Right now, I'm working with people with Alphas and PPC machines (I just got a loaner from a good friend of the Open Source community a couple of weeks back and I've just got around to putting SuSE on the box- PPC support is very likely to become a reality.) The Voodoo/Glide setup is notably harder and much twitchier- I know, I've GOT a Voodoo3; it's because of the implementation, not OpenGL/Mesa.
Some of the people out there contacted them to see what they had to say about all of this. They're supposedly waiting and seeing what comes of all of this. Another submarine attack, perhaps. But, if you look at what I scrounged up, even NeoMedia doesn't have a leg to stand on- the patent DOES cover this sort of thing.
The following is a copy of an e-mail I sent to FBM and the author of FooCat. It is in response to something that was, up until yesterday, on his site, quoting an interview the DC CEO had with an east-coast newspaper. In the interview, the CEO mentions that it's a patent that they acquired that was issued by the USPTO in 1991, that purportedly covers scanning a barcode and generating a network event. If the interview is accurate and doesn't have typos or incorrect info, the following information would be of import (Note: This is supposition. It means nothing until DC fesses up to what IP FooCat and others are violating their rights on- this looks close enough for them to bluff their way through this with..):
A quick search of the Delphion IPN (Used to be IBM's patent site, but they merged their operations with Delphion...) for "(bar code) and (network)" produced only 3 hits for patents in 1991:
#4,982,346 - Mall promotion network apparatus and method
#5,029,183 - Packet data communication network
#5,029,034 - Video casette with optical output of information
A computer system automates advertising and promotional campaigns. The computer system includes a magnetic stripe card reader, bar code reader, monitor, printer, keyboard, and touchscreen input device. Software executing on the computer manages the operations of these devices. The system displays advertisements and product or store locator maps, dispenses coupons, accepts product orders, and manages customer surveys. Customers are attracted to the system by promotional sweepstakes, thereby enhancing the effectiveness of the advertising and surveys. A frequent shopper campaign also attracts customers to the system.
This does NOT cover scanning a barcode and generating a network event. The patent copy is one of the sillier things I've had the misfortune of seeing- 76(!) pages, mostly of flow charts describing the high-level details of the application. It might be said that this is an applicable patent because of what they're claiming, but what they're claiming is not novel, nor is what they're claiming a specific to the WWW or to the Internet in general.
According to Delphion, the current owner of the patent is Inter-Act Systems, Inc. of Norwalk, Conn. which was a change of hands on June 6, 1997. Inter-Act is still in business, doing the same line of business, mall kiosks. Now,
either they've purchased the rights to this patent, they're a subsidiary of Inter-Act (Not likely), they don't own the rights to this patent and they're bluffing because they found it just like I did and they're using it as a smokescreen, or we're barking up the wrong tree. Which is it? Your guess is as good as mine, but they're not
listed as an assignee to the best of what I can find out.
The next closest is #5,029,183. This describes a patent owned by Symbol that is used for remote, hand-held, bar code scanning units to collect data for a central computer (Like those nifty hand-held units that Wal-Mart, K-Mart, etc. use for inventory, etc.). In this case, they'd _have_ to have just bought the patent rights on this one. Symbol's NOT going to relinquish such an important
patent (even if it is overbroad and non-unique...).
Either we've got the date wrong or he's mis-applying the first patent mentioned.
If this is the patent, and by some perverse twist of fate, Delphion's site info isn't in lockstep with who is the current assignee of the patent, you're going to find that anything of the sort is in violation of the patent. Of course, I may have missed the patent (There's something buried, not in the abstracts...)- having said this, I'd like to point out that without it being in the abstracts, it's not likely to be a valid use of the patent grant, because they've got to specifically mention what they're patenting there. Furthermore, most of the usages that we see with it that DC's up in arms about isn't really covered by this patent- the usages don't link up with any marketing data except the case of the engines flipping you to Amazon, and that's tenuous too.
There's kernel "forks" for hard (deterministic) real-time (RT-Linux, etc.). There's kernel "forks" for non-MMU machines (ELKS, uCLinux, etc...). So, why not a "fork" for big iron? If the fork for big iron doesn't hinder current modern machines or improves overall operation- it will become the main fork with the one that just supports the older machines becoming like the other "forks" we see today.
Ogg project has an embryonic codec, called Tarkin.
on
DivX ;-) Deux Update
·
· Score: 2
And 3D wavelet transforms are what the codec supposedly uses. It's promising, but they're trying to put the polish on Vorbis before moving to doing serious Tarkin development.
Anyone that doesn't believe that Linux scales, is mistaken. Anyone that professes the same is either ignorant of what's been going down for the past two years or is FUDing.
While I'll admit that the S/390 port is some proof of scalability- every little drop helps dispell myths.
Waited for six weeks while Verizon and Internet America couldn't get their damn act together.... I'm being reliably told by others that the same is the case all over the place.
Cable sucks. xDSL sucks. For different reasons, of course, but it's a hell of a lot better than dialup just the same. Faster and no waiting for the firewall to dial up the ISP.
XORed Base 64 protocol? Barcodes? Hate to tell you this bub, but neither qualify as IP- vast amounts of prior art there, not to mention that nobody that Digital Convergence is harrassing (yes, harrassing) has done anything other than a clean-room reverse engineering of the protocol (which can't be made as IP...) between the barcode reader (which could concievably be IP, but we're not copying the reader- which would be the only possible IP violation in this context) and the PC.
There is no IP or IP violations involved here.
They manufactured the things for DC...
on
CueCat At It Again
·
· Score: 2
Look at the bottom of the things that RS are handing out. It clearly states that Tandy is the manufacturer. They're an investor AND the supplier of their barcode reader- they make money either way it goes down.
Some of the upgrades have been painless. Others, well... My bet's somewhere in-between. As soon as it shows up (and I can get past the/. hordes), I'll pull it down and give it a whirl.
DRI's the answer to that. You're NOT going to go through X with it. And, it seems that with the current RagePRO GLX support through Utah-GLX, that I've got a visually correct display and slightly higher frame rates in many of my games than in Windows. That's through all those layers of indirection that X presents.
Look at all the consoles out there. Of them, which is the only one out there that will have anything resembling it? You guessed it- the X-Box. Of the console choices out there, which one is vaporware? Again, the X-Box (if it's not shipping yet, it's still vapor!). Now, if the console companies seem (and they do at that!) to be doing well for themselves without DirectX, why on this Earth would it be an issue for this machine?
It meets the GL spec in all the areas that matters and it's a decent software implementation. It happens that what I think you're talking about is accelerated MesaGL- which is a different beast altogether. And, yes, without better acceleration support, there's not going to be a future. Thing is, there is a future- DRI is coming. For those that can't wait and have select accelerators, there's Utah-GLX.
Workstations only for work? De facto nature of Linux? I think you might want to take those blinders off. Windows98 and W2k are "workstation" OSes as well (Well, Microsoft calls them that!) and they are used as such. That's the de facto nature of them- but yet, look at all the games for those OSes. Take all the preconcieved notions of what an OS is supposed to be used for and throw them clean out the window. And as for it being "fun" without a high-speed internet connection; you're not trying hard enough.
If you look at the EULA, they claim to own the device as well as the software- that the CueCat reader is on loan to you. That's in violation of the Postal Regs (Esp. if it said "Paid for by Delta Airlines" anywhere on the packaging!).
Yggdrasil is still alive and kicking in the old area they used to operate from. (Yeah, I know...)
Yggdrasil produced a DVD disk of everything on the Linux and GNU software repositories- while you can pull it off of the 'net, and with xDSL and Cable, it's easy to do so, it's nice to have a DVD or CD of a static image for when you've not got a networked situation.
Yggdrasil handed us an application that allows you to make that magic DLT formatting that a pressing house will only accept to make production DVDs.
They've been quietly chugging away, working consulting gigs up until recently. I think they might have thought that their distribution was more work than it was making them money and discontinued the releases except for subscription users.
They send the info to a central repository in Ft. Worth at One Tandy Center. But, there's no mojo in the system that I know of that knows that they're getting duplicates (Nor, do I think they care they care- they're getting their money no matter what; they MADE the silly readers for DC!) as the system there doesn't care and the locally collected info isn't replicated to the other stores (think about that for a moment; they don't want that much data and couldn't manage that much locally!)
You're talking about mastering and burning a DVD recordable disk. Those are single layer affairs and can only hold 4Gb- end of story. This is a production double layer disk produced by a pressing house. The pressing houses only accept DLTs with a special formatting of the data- previously, you needed expensive, proprietary applications (that ran on Windows, for the most part, from what I understand) to master the tape for pressing. The Yggdrasil app does the tape mastering for you.
Writable DVDs are single layer- meaning only 4 Gb.
Production DVDs are double layer and the pressing houses want the data fed to them in a just-so format on a DLT. What Yggdrasil has done is make a program to master that tape for you without needing a Windows box and expensive commercial software. For a DVD writer, I suspect that the traditional tools work just fine.
NFS is grotesquely insecure. If you have to use NFS, use it behind a seriously locked down firewall box. If you can avoid using it, use anything other than it. SMB is also less desirable than others but it's design won't leave you open for these attacks. I suggest using AFS or Coda at this point.
Slashdot Mirror
Speed is obtained by doing it for yourself down to the barest Win32 API layers. Power is obtained the same way. MFC buys one thing- laziness. It's easy to bang out a UI for an office/buisness app with it. Try to do something fancy with it- forget it.
WINE derived apps are slower than many native ones. Why? Because it's translating things over to native behaviors. It's still a translation layer, albeit a thin one.
Which would I rather have? If I had a choice of no app or a WINE app? Dumb question. If I had the choice of a native Qt, Fltk, Wx, JX, GTK+, etc. app versus a WINE app? Dumb question, again- native is better than non-native.
If the game's got multiple threads or has a seperate server daemon process, the SMP machine will edge out the uni-processor one if the SMP machine's CPUs are anything faster than half the uni-processor's CPU. Same goes for anything out there that has threads or has external processing applications. I don't want a 1GHz machine- I'm perfectly happy with a dual or quad 750 for now.
Greetings! I couldn't remember which big paper it was- Thanks!
Keep plugging away at these clowns- they deserve every drop of agony that's about to come to them.
If you've got a RagePRO, it's merely a matter of adding a "Module" section with a 'Load "glx.so"' in it, copying/linking 4 files in the right places (Three files/links go to /usr/X11R6/lib- all the libGL* files. One file goes in /usr/X11R6/lib/modules- glx.so.) and you're good to go with Utah-GLX.
It is literally that simple.
Right now, I'm working with people with Alphas and PPC machines (I just got a loaner from a good friend of the Open Source community a couple of weeks back and I've just got around to putting SuSE on the box- PPC support is very likely to become a reality.) The Voodoo/Glide setup is notably harder and much twitchier- I know, I've GOT a Voodoo3; it's because of the implementation, not OpenGL/Mesa.
Some of the people out there contacted them to see what they had to say about all of this. They're supposedly waiting and seeing what comes of all of this. Another submarine attack, perhaps. But, if you look at what I scrounged up, even NeoMedia doesn't have a leg to stand on- the patent DOES cover this sort of thing.
If this is the patent, and by some perverse twist of fate, Delphion's site info isn't in lockstep with who is the current assignee of the patent, you're going to find that anything of the sort is in violation of the patent. Of course, I may have missed the patent (There's something buried, not in the abstracts...)- having said this, I'd like to point out that without it being in the abstracts, it's not likely to be a valid use of the patent grant, because they've got to specifically mention what they're patenting there. Furthermore, most of the usages that we see with it that DC's up in arms about isn't really covered by this patent- the usages don't link up with any marketing data except the case of the engines flipping you to Amazon, and that's tenuous too.
There's kernel "forks" for hard (deterministic) real-time (RT-Linux, etc.). There's kernel "forks" for non-MMU machines (ELKS, uCLinux, etc...). So, why not a "fork" for big iron? If the fork for big iron doesn't hinder current modern machines or improves overall operation- it will become the main fork with the one that just supports the older machines becoming like the other "forks" we see today.
And 3D wavelet transforms are what the codec supposedly uses. It's promising, but they're trying to put the polish on Vorbis before moving to doing serious Tarkin development.
Anyone that doesn't believe that Linux scales, is mistaken. Anyone that professes the same is either ignorant of what's been going down for the past two years or is FUDing.
While I'll admit that the S/390 port is some proof of scalability- every little drop helps dispell myths.
Bandwidth is as expected. Throughput the same.
Waited for six weeks while Verizon and Internet America couldn't get their damn act together.... I'm being reliably told by others that the same is the case all over the place.
Cable sucks. xDSL sucks. For different reasons, of course, but it's a hell of a lot better than dialup just the same. Faster and no waiting for the firewall to dial up the ISP.
XORed Base 64 protocol? Barcodes? Hate to tell you this bub, but neither qualify as IP- vast amounts of prior art there, not to mention that nobody that Digital Convergence is harrassing (yes, harrassing) has done anything other than a clean-room reverse engineering of the protocol (which can't be made as IP...) between the barcode reader (which could concievably be IP, but we're not copying the reader- which would be the only possible IP violation in this context) and the PC.
There is no IP or IP violations involved here.
Look at the bottom of the things that RS are handing out. It clearly states that Tandy is the manufacturer. They're an investor AND the supplier of their barcode reader- they make money either way it goes down.
Some of the upgrades have been painless. Others, well... My bet's somewhere in-between. As soon as it shows up (and I can get past the /. hordes), I'll pull it down and give it a whirl.
DRI's the answer to that. You're NOT going to go through X with it. And, it seems that with the current RagePRO GLX support through Utah-GLX, that I've got a visually correct display and slightly higher frame rates in many of my games than in Windows. That's through all those layers of indirection that X presents.
DirectX is not a magic bullet. DirectX is not a way to make Linux popular as a platform for gaming.
Look at all the consoles out there. Of them, which is the only one out there that will have anything resembling it? You guessed it- the X-Box. Of the console choices out there, which one is vaporware? Again, the X-Box (if it's not shipping yet, it's still vapor!). Now, if the console companies seem (and they do at that!) to be doing well for themselves without DirectX, why on this Earth would it be an issue for this machine?
It meets the GL spec in all the areas that matters and it's a decent software implementation. It happens that what I think you're talking about is accelerated MesaGL- which is a different beast altogether. And, yes, without better acceleration support, there's not going to be a future. Thing is, there is a future- DRI is coming. For those that can't wait and have select accelerators, there's Utah-GLX.
Workstations only for work? De facto nature of Linux? I think you might want to take those blinders off. Windows98 and W2k are "workstation" OSes as well (Well, Microsoft calls them that!) and they are used as such. That's the de facto nature of them- but yet, look at all the games for those OSes. Take all the preconcieved notions of what an OS is supposed to be used for and throw them clean out the window. And as for it being "fun" without a high-speed internet connection; you're not trying hard enough.
If you look at the EULA, they claim to own the device as well as the software- that the CueCat reader is on loan to you. That's in violation of the Postal Regs (Esp. if it said "Paid for by Delta Airlines" anywhere on the packaging!).
Yggdrasil is still alive and kicking in the old area they used to operate from. (Yeah, I know...)
Yggdrasil produced a DVD disk of everything on the Linux and GNU software repositories- while you can pull it off of the 'net, and with xDSL and Cable, it's easy to do so, it's nice to have a DVD or CD of a static image for when you've not got a networked situation.
Yggdrasil handed us an application that allows you to make that magic DLT formatting that a pressing house will only accept to make production DVDs.
They've been quietly chugging away, working consulting gigs up until recently. I think they might have thought that their distribution was more work than it was making them money and discontinued the releases except for subscription users.
They send the info to a central repository in Ft. Worth at One Tandy Center. But, there's no mojo in the system that I know of that knows that they're getting duplicates (Nor, do I think they care they care- they're getting their money no matter what; they MADE the silly readers for DC!) as the system there doesn't care and the locally collected info isn't replicated to the other stores (think about that for a moment; they don't want that much data and couldn't manage that much locally!)
You're talking about mastering and burning a DVD recordable disk. Those are single layer affairs and can only hold 4Gb- end of story. This is a production double layer disk produced by a pressing house. The pressing houses only accept DLTs with a special formatting of the data- previously, you needed expensive, proprietary applications (that ran on Windows, for the most part, from what I understand) to master the tape for pressing. The Yggdrasil app does the tape mastering for you.
Writable DVDs are single layer- meaning only 4 Gb.
Production DVDs are double layer and the pressing houses want the data fed to them in a just-so format on a DLT. What Yggdrasil has done is make a program to master that tape for you without needing a Windows box and expensive commercial software. For a DVD writer, I suspect that the traditional tools work just fine.
NFS is grotesquely insecure. If you have to use NFS, use it behind a seriously locked down firewall box. If you can avoid using it, use anything other than it. SMB is also less desirable than others but it's design won't leave you open for these attacks. I suggest using AFS or Coda at this point.