* Their graphics cards would become invaluable for rendering production output for TV, film and video
Oh yeah, the current market for this is huuuge, NOT!
When and if the need arises I'm sure the card manufacturers will address this. Right now it's FPS that count and I don't think GFX companies are going to waste engineering time on this useless feature without any significant return on investment.
* Users could actually record game output in real-time with little impact on game performance. After playing there could be a compressed movie of your game play saved on your HD. On a reasonably fast machine you could actually record your game play digitally to your DV camcorder as you play or even compress and burn it to a Video CD or DVD at the same time you are actually playing
This one is just silly. Why not record the game engine commands instead of the videocard output? Oh wait, that has been possible for years in most FPS games no? What these people are proposing is to capture high resolution images, compress (eeks), and them stream out to a TV screen that probably has only 1/3 the resolution of the original capture. What a great way to waste time!
* Screen capture software that grabs motion images of user interfaces for the purposes of tutorials and training is a vital business application
Haha, this one takes the cake. Most of these tutorials do not need high FPS numbers to be usefull at all. And even more importantly, a lot of these applications simply script the real application to demonstrate the needed behaviour. You can't beat that.
* Despite the popularity of Internet streaming, it is not currently possible to stream live output from graphics cards over the Internet. The connections, processors and codecs are all fast enough today. Sadly, all of this horsepower is being held back by one remaining weak link: the texture download speed of today's graphics card drivers.
Woehahhahaa:-)
I'd rather fetch a UDP stream of game engine commands render the game action on my side of the Internet, thank you very much.
What a joke:)
-adnans
Re:OH WELL! I'm an idiot! I don't know ****!
on
VisionTek Folds
·
· Score: 2
Show me where Nvidia has helped with the creation of open source drivers
Okay, so I'm shopping for a new videocard. First a layout of some observations and requirements:
1) I want the most bang for my (300,-) euros
2) No matter which card I buy, it will be obsolete in 2 years time (my next upgrade)
3) It should run flawlessly in Linux, meaning both 2D, 3D and TV-out should be excellent and at fullspeed.
4) It must be Doom III ready:)
First let's check out Matrox, the other famous "Open Source" supporting chipset/card manufacturer. Parhelia drivers are not available at this time ("Coming Soon", that could be tomorrow, or next year), so scratch the Parhelia. I do have a G400 MAX I bought 3 years ago, hmmmm. The TV-out support requires a closed-source binary module, so they're not much better than NVidia when it comes to Open Source. On top of that, RENDER support has been utterly broken on TV-out for well over a year. Oh and you do not get hardware accellerated OpenGL or Xv(ideo) support on the second head. The G550 suffers the same braindeadness BTW. With a track record like that Matrox is definitely not an option right now.
Next ATI. Radeon 9700, sweeet! However there's no news on wether or not we'll have full Linux support for this beast, and when! The 8500 is supposed to be fully supported by the end of 2002 (when it will be about 2 years old!). Let's wait and see...perhaps at the next upgrade cycle.
And finally NVidia. Let's see, unified driver architecture, so both Linux and Windows support are on-par. The best Linux OpenGL stack at this time (ask John Carmack). Excellent Xv support. The main Xv architect is on NVidia's payroll BTW. He's also one of the most active X guru's these days, currently coding the new XAA architecture. Very good TV-out with accellerated RENDER *and* OpenGL support. Definitely the best of the flock!
So today I have a brandspanking new Geforce4 ti4400 128MB which replaced my aging Geforce2 GTS. Updating was as simple as swiping the cards.
I'm a strong Open Source supporter, but there is one area where Open Source hasn't proven itself yet: up-to-date videocard drivers...
the only thing the server does is draw trapezoids (or triangles). Render is based on image compositing so it provides translucent images as well as anti-aliased objects.
How far off are we to getting these basic operations hardware accellerated? I.e. what needs to be done still?
For real translucancy I read that we need another X extension. Isn't it possible to draw each toplevel X window in an offscreen buffer (a la Quartz Extreme) and then use XRENDER to composite all these windows to the final root window?
If you put loops into the language, does this present a nice denial of service approach?
I doubt you could force infinite loops in the drawing engine from any application. Clients are after all scheduled on the X server, much like regular processes are scheduled on the CPU. Of course if your client has a higher priority than all the other drawing clients you could effectively lock them out, but that implies having enough rights to gain lower niceness levels anyways.
What I'd really like to know if this new drawing engine will allow us to do similar thing like Quartz Extreme. I would love to use the power that's locked into my Geforce4 for regular 2D (read: xterm) needs:-) My stomach always turns when I see l33t Linux pseudo transparant xterm's and KDE/GNOME widgets in screenshots...
It is for me. And seeing how it defeats about 90% of ALL incoming SPAM I think I'll keep that requirement. If you don't take the time to properly configure reverse-DNS lookup, if only for your SMTP server, I simply bounce your message with an appropiate 50X reply. If you can't live with that, tough!
Read the related RFC's
LOL. What the hell does an RFC have to do with combatting SPAM at my doorstep?
Anyone who bounces mail based on a lack of a reverse lookup has no concept of how to properly run a host.
don't you hate listening to the breaks in the music between tracks 4 and 5 of pink floyd's
I have some live albums (including Pink Floyd's P.U.L.S.E.) and there is no perceived gap between songs. You can further tweek the 1/10th second startup time of a new song in the playlist. The only requirement is that your storage media is fast enough to not loose to much time finding the next file.
i'd love to use a different engine, but it needs a gapkiller plugin.
I will think about it, it should be fairly easy to implement. One advantage AlsaPlayer has is that it can do it's own internal mixing i.e. you don't need to do funny buffering to do stuff like crossfading and I would think gap detection. This keeps the controls very responsive because there's no need for very deep audio buffers.
Consider using AlsaPlayer as the backend for your project. AlsaPlayer supports so called interface plugins, where you can write your own custom front-end to the player if your needs are that specific. The current CVS version
supports a "daemon" interface where the player will just run as a background process and accept commands through the
libalsaplayer control interface. There are already a couple of projects preparing to switch to this interface. I know of at least one commercial project that is currently programming a backend.
I prefer to block SPAM at the door with Exim. This does mean that you will have to your own MX hosting which might be more trouble than it's worth for most folks. It also requires mail filtering since you typically detect a source of spam when one ore more messages get through. After 3 months of monitoring I have collected a number of hostmask (reverse-DNS challenged hosts are automatically blocked). This collection manages to block out about 98% of all SPAM messages.. during the SMTP handshake! The remaining 2% are new messages from (mostly open Windows NT) relays that do get blocked by the other filters. Some of the more notorious and persistent sources:
Today's addition: *.passionup.com:) Given, there are still a couple of messages that pass through all the filters, but the last such message I got was about 2 weeks ago. Not fullproof, but getting htere...
Never. BeOS is dead. Perhaps you'll see parts of the API in PalmOS 5.x, but there won't be any new release of BeOS. IF you want a BeOS like enviroment check out Cosmoe OS (it runs on top of Linux).
I'll take unified (preferably open source) drivers over Free Software drivers this time. My G400 was already obsolete for a good year or two when semi usable free software drivers for it emerged. The reality is that people capable of writing good free drivers are in short supply.
I'd rather have all Matrox driver engineers slave away at good drivers for my Linux box than one or two (unpaid?) engineers volunteering their spare time. Anything else would basically mean wasting good money on hardware that I can't really use.
Only if your Linux system supports the old a.out exec format and the ancient libc installed, no?!
Go ahead, grab
XF86-2.1-bin.tar.gz and see if any of the binaries run:-)
It is smoother than any other "native" X toolkit I'v see so far. I'm guessing it is coded on top of OpenGL. I think there a much bigger market for that kind of application:)
Except I didn't paint it black. Seriously, everyone geeky enough to be interested in this box can probably build his/her own and much cheaper! And why not throw in:
It depends;) Using proper drivers (NVIDIA, which allows the use of XVideo extensions)
Unfortunately the Xv extension only allows one application to open a port at a time, so only the first movie will benefit from Xv hardware scaling for example. However, this is a hardware limitation and not X's fault.
Side-note: Thanks Andy for the awesome BeMP. I used that mp3 player more than I can remember
Me too! *g* Great to hear that it was useful to other folks too!!
I have to say though, some of the code was nasty.
BeMP was actually my first (and last) BeOS GUI app so the code was basically a collection of hacks, experiments and discoveries all massaged in such a way that it still compiled:)
I rewrote a good portion of it before I gave up and created my own personal BeMP, then promptly stopped using BeOS.
FinalScratch runs on Linux now, since BeOS is pretty much dead. At least that's what I heard.
You can simulate a cheap Finalscratch by using AlsaPlayer and feeding the speed parameter via an external program, using libalsaplayer. Some folks are already doing this in a lab. There's this rumour that FinalScratch is using some bits of AlsaPlayer (plugins), have to check if this is really true;)
Actually, none of those patches at this time. I do have a ll kernel handy for testing, but it has its problems (particularly in the disk I/O area, and VMware breaks with ll for now). Those things were done with "vanilla" kernel 2.4.19-pre3:-)
My el cheapo trident soundcard can do sustained 1.3ms max latency audio output for hours on end with the ll patches. Not in full duplex mode though, but that's a hardware limitation (read: brokeness).
So can you open 42 versions of the same.avi movie on your Linux Desktop and play them simultaneously without dropped frames while surfing the internet on a PII450?
Yes! Unfortunately (or fortunately) I don't have a PII450. BTW, most of those "cool" demos of BeOS running tens of player windows are done with a single avi (worthless to to test the real I/O), very low resolution, and without sound.
Not to diss Linux or anything, but I don't think it is time to proclaim Linux has eclipsed Be's technology.
You're right, that time has come and passed us! I'm playing 9 *different* avi's now (served over 100Mbit NFS btw), 3 mp3's *backwards* (haha, another stupid benchmark thrown in:), with xawtv in a corner, and typing this message in mozilla 0.9.9, and I still have CPU cycles to spare! Linux rocks;-)
Are you scheduling your MP3 player at a higher prioritary (*) than say, your cp command? It is very important that you do this. While this might not fix the massive skipping it could improve things a great deal. Given, there are still lots of problems with the disk I/O subsystem in Linux that neither the preempt or ll patches fix right now so the occasional skip under heavy load is guaranteed, especially since consumer grade audiocards are not really catering towards low latency.
<plug>
AlsaPlayer comes with a --realtime switch that enabled SCHED_FIFO scheduling for the audio thread, this eliminates skipping for most people, but requires the binary to be run as root (or be suid root)
</plug>
-adnans
(*) scheduling something at a higher prio usually requires running the program suid root, or
calling renice(8) as root with the programs pid. A better method is to sudo the nice command for the given user so he/she can nice the app at the start
I don't know what's wrong with the mounting issue, but what kind of faster algorithm is he talking about here?
He is a she, and like all (ex-)BeOS users, Eugenia is
obsessed with bootup speed, because BeOS booted in about 12 seconds on most machines (it actually booted in 8 seconds on my BeBox during the DR days). A small amount of tweaking got me a 15 second bootup time in Linux (from the GRUB prompt to my X session manager, xdm), and that includes SCSI initialization. The main rule is of course is to avoid anything that starts with a K!:) But then again, I only reboot when upgrading kernels so...
However, in all my Linux years I have never ever waited *2 minutes* for a partition to mount. So either Eugenia's box is terribly slow, or something is very wrong with that Linux distro, or she's lying:)
Whoops, sorry, SGI was there first with XFS. BeFS was actually modelled after XFS. These days Linux has XFS with all its kick-arse features too, including extended attributes, and options that never were in BeFS (quotas, real-time I/O, NFS, etc) . The only thing missing is the automatic attribute indexing, but that could very well be implemented in userspace if you use the node monitoring feature of XFS. Come to think of it, that would be a damn cool project to do:)
I have never heard a customer complain about speed - I assume that they blame any speed deficiencies on their hardware. So where is the problem - and incentive for us to improve speed ?
It really depends on what kind of optimization this is. In theory having a compiler that produces optimised code is good no matter what, especially in the case of Altivec, because it is assumed that faster processors will also have the Altivec instruction set. It's a win win situation: optimise for Altivec.. speed boost, faster processor with Altivec comes out.. speed boost!
Definitely worth it I think!
-adnans
Now we only need...
on
GTK+ 2.0
·
· Score: 0, Troll
a Mozilla that doesn't rubberband like hell, and doesn't bloody BLOCK all freaking windows when one window is busy:(
Slashdot Mirror
* Their graphics cards would become invaluable for rendering production output for TV, film and video
:-)
:)
Oh yeah, the current market for this is huuuge, NOT! When and if the need arises I'm sure the card manufacturers will address this. Right now it's FPS that count and I don't think GFX companies are going to waste engineering time on this useless feature without any significant return on investment.
* Users could actually record game output in real-time with little impact on game performance. After playing there could be a compressed movie of your game play saved on your HD. On a reasonably fast machine you could actually record your game play digitally to your DV camcorder as you play or even compress and burn it to a Video CD or DVD at the same time you are actually playing
This one is just silly. Why not record the game engine commands instead of the videocard output? Oh wait, that has been possible for years in most FPS games no? What these people are proposing is to capture high resolution images, compress (eeks), and them stream out to a TV screen that probably has only 1/3 the resolution of the original capture. What a great way to waste time!
* Screen capture software that grabs motion images of user interfaces for the purposes of tutorials and training is a vital business application
Haha, this one takes the cake. Most of these tutorials do not need high FPS numbers to be usefull at all. And even more importantly, a lot of these applications simply script the real application to demonstrate the needed behaviour. You can't beat that.
* Despite the popularity of Internet streaming, it is not currently possible to stream live output from graphics cards over the Internet. The connections, processors and codecs are all fast enough today. Sadly, all of this horsepower is being held back by one remaining weak link: the texture download speed of today's graphics card drivers.
Woehahhahaa
I'd rather fetch a UDP stream of game engine commands render the game action on my side of the Internet, thank you very much.
What a joke
-adnans
Show me where Nvidia has helped with the creation of open source drivers
:)
Okay, so I'm shopping for a new videocard. First a layout of some observations and requirements:
1) I want the most bang for my (300,-) euros
2) No matter which card I buy, it will be obsolete in 2 years time (my next upgrade)
3) It should run flawlessly in Linux, meaning both 2D, 3D and TV-out should be excellent and at fullspeed.
4) It must be Doom III ready
First let's check out Matrox, the other famous "Open Source" supporting chipset/card manufacturer. Parhelia drivers are not available at this time ("Coming Soon", that could be tomorrow, or next year), so scratch the Parhelia. I do have a G400 MAX I bought 3 years ago, hmmmm. The TV-out support requires a closed-source binary module, so they're not much better than NVidia when it comes to Open Source. On top of that, RENDER support has been utterly broken on TV-out for well over a year. Oh and you do not get hardware accellerated OpenGL or Xv(ideo) support on the second head. The G550 suffers the same braindeadness BTW. With a track record like that Matrox is definitely not an option right now.
Next ATI. Radeon 9700, sweeet! However there's no news on wether or not we'll have full Linux support for this beast, and when! The 8500 is supposed to be fully supported by the end of 2002 (when it will be about 2 years old!). Let's wait and see...perhaps at the next upgrade cycle.
And finally NVidia. Let's see, unified driver architecture, so both Linux and Windows support are on-par. The best Linux OpenGL stack at this time (ask John Carmack). Excellent Xv support. The main Xv architect is on NVidia's payroll BTW. He's also one of the most active X guru's these days, currently coding the new XAA architecture. Very good TV-out with accellerated RENDER *and* OpenGL support. Definitely the best of the flock!
So today I have a brandspanking new Geforce4 ti4400 128MB which replaced my aging Geforce2 GTS. Updating was as simple as swiping the cards.
I'm a strong Open Source supporter, but there is one area where Open Source hasn't proven itself yet: up-to-date videocard drivers...
-adnans
the only thing the server does is draw trapezoids (or triangles). Render is based on image compositing so it provides translucent images as well as anti-aliased objects.
How far off are we to getting these basic operations hardware accellerated? I.e. what needs to be done still? For real translucancy I read that we need another X extension. Isn't it possible to draw each toplevel X window in an offscreen buffer (a la Quartz Extreme) and then use XRENDER to composite all these windows to the final root window?
Thanks...
-adnans
If you put loops into the language, does this present a nice denial of service approach?
:-) My stomach always turns when I see l33t Linux pseudo transparant xterm's and KDE/GNOME widgets in screenshots...
I doubt you could force infinite loops in the drawing engine from any application. Clients are after all scheduled on the X server, much like regular processes are scheduled on the CPU. Of course if your client has a higher priority than all the other drawing clients you could effectively lock them out, but that implies having enough rights to gain lower niceness levels anyways.
What I'd really like to know if this new drawing engine will allow us to do similar thing like Quartz Extreme. I would love to use the power that's locked into my Geforce4 for regular 2D (read: xterm) needs
-adnans
Reverse DNS is NOT A REQUIREMENT TO HOST MAIL
:)
It is for me. And seeing how it defeats about 90% of ALL incoming SPAM I think I'll keep that requirement. If you don't take the time to properly configure reverse-DNS lookup, if only for your SMTP server, I simply bounce your message with an appropiate 50X reply. If you can't live with that, tough!
Read the related RFC's
LOL. What the hell does an RFC have to do with combatting SPAM at my doorstep?
Anyone who bounces mail based on a lack of a reverse lookup has no concept of how to properly run a host.
Riiiiight!
-adnans
don't you hate listening to the breaks in the music between tracks 4 and 5 of pink floyd's
I have some live albums (including Pink Floyd's P.U.L.S.E.) and there is no perceived gap between songs. You can further tweek the 1/10th second startup time of a new song in the playlist. The only requirement is that your storage media is fast enough to not loose to much time finding the next file.
i'd love to use a different engine, but it needs a gapkiller plugin.
I will think about it, it should be fairly easy to implement. One advantage AlsaPlayer has is that it can do it's own internal mixing i.e. you don't need to do funny buffering to do stuff like crossfading and I would think gap detection. This keeps the controls very responsive because there's no need for very deep audio buffers.
-adnans
<plug>
Consider using AlsaPlayer as the backend for your project. AlsaPlayer supports so called interface plugins, where you can write your own custom front-end to the player if your needs are that specific. The current CVS version supports a "daemon" interface where the player will just run as a background process and accept commands through the libalsaplayer control interface. There are already a couple of projects preparing to switch to this interface. I know of at least one commercial project that is currently programming a backend.
</plug>
-adnans
I prefer to block SPAM at the door with Exim. This does mean that you will have to your own MX hosting which might be more trouble than it's worth for most folks. It also requires mail filtering since you typically detect a source of spam when one ore more messages get through. After 3 months of monitoring I have collected a number of hostmask (reverse-DNS challenged hosts are automatically blocked). This collection manages to block out about 98% of all SPAM messages.. during the SMTP handshake! The remaining 2% are new messages from (mostly open Windows NT) relays that do get blocked by the other filters. Some of the more notorious and persistent sources:
:) Given, there are still a couple of messages that pass through all the filters, but the last such message I got was about 2 weeks ago. Not fullproof, but getting htere...
*.pacbell.net
*.mb0?.net
*.client.dsl.net
*.dsl.att.net
*.attbi.com
Today's addition: *.passionup.com
-adnans
When are we going to see a new release of BeOS?
Never. BeOS is dead. Perhaps you'll see parts of the API in PalmOS 5.x, but there won't be any new release of BeOS. IF you want a BeOS like enviroment check out Cosmoe OS (it runs on top of Linux).
-adnans
This has got to be one of the funniest stories I've ever read on /. :-)
-adnans
I'll take unified (preferably open source) drivers over Free Software drivers this time. My G400 was already obsolete for a good year or two when semi usable free software drivers for it emerged. The reality is that people capable of writing good free drivers are in short supply.
:)
I'd rather have all Matrox driver engineers slave away at good drivers for my Linux box than one or two (unpaid?) engineers volunteering their spare time. Anything else would basically mean wasting good money on hardware that I can't really use.
-adnans (not an NVidia pimp
Only if your Linux system supports the old a.out exec format and the ancient libc installed, no?! :-)
/tmp/bin$ file xload
Go ahead, grab XF86-2.1-bin.tar.gz and see if any of the binaries run
xload: Linux/i386 demand-paged executable ZMAGIC), stripped
-adnans
It is smoother than any other "native" X toolkit I'v see so far. I'm guessing it is coded on top of OpenGL. I think there a much bigger market for that kind of application :)
-adnans
and
All that in a box almost half the price?!
-adnans
P.S. the pictures are very old, should take some new ones
It depends ;) Using proper drivers (NVIDIA, which allows the use of XVideo extensions)
Unfortunately the Xv extension only allows one application to open a port at a time, so only the first movie will benefit from Xv hardware scaling for example. However, this is a hardware limitation and not X's fault.
-adnans
Side-note: Thanks Andy for the awesome BeMP. I used that mp3 player more than I can remember
:)
:(
Me too! *g* Great to hear that it was useful to other folks too!!
I have to say though, some of the code was nasty.
BeMP was actually my first (and last) BeOS GUI app so the code was basically a collection of hacks, experiments and discoveries all massaged in such a way that it still compiled
I rewrote a good portion of it before I gave up and created my own personal BeMP, then promptly stopped using BeOS.
I actually stopped using BeOS in the R4 days
Thanks for the nostalgia trip!
-adnans
FinalScratch runs on Linux now, since BeOS is pretty much dead. At least that's what I heard.
;)
You can simulate a cheap Finalscratch by using AlsaPlayer and feeding the speed parameter via an external program, using libalsaplayer. Some folks are already doing this in a lab. There's this rumour that FinalScratch is using some bits of AlsaPlayer (plugins), have to check if this is really true
-adnans
Actually, none of those patches at this time. I do have a ll kernel handy for testing, but it has its problems (particularly in the disk I/O area, and VMware breaks with ll for now). Those things were done with "vanilla" kernel 2.4.19-pre3 :-)
My el cheapo trident soundcard can do sustained 1.3ms max latency audio output for hours on end with the ll patches. Not in full duplex mode though, but that's a hardware limitation (read: brokeness).
-adnans
So can you open 42 versions of the same .avi movie on your Linux Desktop and play them simultaneously without dropped frames while surfing the internet on a PII450?
:), with xawtv in a corner, and typing this message in mozilla 0.9.9, and I still have CPU cycles to spare! Linux rocks ;-)
Yes! Unfortunately (or fortunately) I don't have a PII450. BTW, most of those "cool" demos of BeOS running tens of player windows are done with a single avi (worthless to to test the real I/O), very low resolution, and without sound.
Not to diss Linux or anything, but I don't think it is time to proclaim Linux has eclipsed Be's technology.
You're right, that time has come and passed us! I'm playing 9 *different* avi's now (served over 100Mbit NFS btw), 3 mp3's *backwards* (haha, another stupid benchmark thrown in
-adnans (ex-BeOS user)
Are you scheduling your MP3 player at a higher prioritary (*) than say, your cp command? It is very important that you do this. While this might not fix the massive skipping it could improve things a great deal. Given, there are still lots of problems with the disk I/O subsystem in Linux that neither the preempt or ll patches fix right now so the occasional skip under heavy load is guaranteed, especially since consumer grade audiocards are not really catering towards low latency.
<plug>
AlsaPlayer comes with a --realtime switch that enabled SCHED_FIFO scheduling for the audio thread, this eliminates skipping for most people, but requires the binary to be run as root (or be suid root)
</plug>
-adnans
(*) scheduling something at a higher prio usually requires running the program suid root, or calling renice(8) as root with the programs pid. A better method is to sudo the nice command for the given user so he/she can nice the app at the start
When will those Dune windtraps become reality??
Seriously, power generation via wave is old news.
Check out this site for some backgrounds.
-adnans
I don't know what's wrong with the mounting issue, but what kind of faster algorithm is he talking about here?
:) But then again, I only reboot when upgrading kernels so...
:)
He is a she, and like all (ex-)BeOS users, Eugenia is obsessed with bootup speed, because BeOS booted in about 12 seconds on most machines (it actually booted in 8 seconds on my BeBox during the DR days). A small amount of tweaking got me a 15 second bootup time in Linux (from the GRUB prompt to my X session manager, xdm), and that includes SCSI initialization. The main rule is of course is to avoid anything that starts with a K!
However, in all my Linux years I have never ever waited *2 minutes* for a partition to mount. So either Eugenia's box is terribly slow, or something is very wrong with that Linux distro, or she's lying
-adnans
Be had the best FS
:)
Whoops, sorry, SGI was there first with XFS. BeFS was actually modelled after XFS. These days Linux has XFS with all its kick-arse features too, including extended attributes, and options that never were in BeFS (quotas, real-time I/O, NFS, etc) . The only thing missing is the automatic attribute indexing, but that could very well be implemented in userspace if you use the node monitoring feature of XFS. Come to think of it, that would be a damn cool project to do
-adnans
I have never heard a customer complain about speed - I assume that they blame any speed deficiencies on their hardware. So where is the problem - and incentive for us to improve speed ?
.. speed boost, faster processor with Altivec comes out .. speed boost!
It really depends on what kind of optimization this is. In theory having a compiler that produces optimised code is good no matter what, especially in the case of Altivec, because it is assumed that faster processors will also have the Altivec instruction set. It's a win win situation: optimise for Altivec
Definitely worth it I think!
-adnans
a Mozilla that doesn't rubberband like hell, and doesn't bloody BLOCK all freaking windows when one window is busy :(