The.pdf didn't indicate (it's a PR/marketing piece,) so I'm assuming that the new extensions will be contributed to the OpenGL folks for inclusion. Is this correct? If so, a very nice contribution by the Apple folks!
The wording is a little deceptive.
Most of these are extensions already existing for Windows and other OpenGL ports. NVidia example. ATI example. What they're basically saying is that the Mac drivers are caught up, and/or use of the extension is new to Jaguar's version of the Quartz engine.
There are a few Apple-specific extensions in there, but they're very specially purposed to Quartz' preferred data formats. Essentially, they're just a way to reduce the portability of the system (restricted pixel formats) in favor of some speed boosts, which is a pretty fair tradeoff if you're a company like Apple who only deals with a pocketful of vendors who make special concessions. You wouldn't want these back in OpenGL main.
There's some damned fine engineering going on at Apple, as always. But there's also the familiar nice spin, though. I wish they'd keep that much out of the technical presentations, or at least would more clearly mark it as such.
There's a system called "Blow" which allows morse-code-like typing through a straw.
A previous employer told me that Shanghai was written by a disabled programmer, using Blow. I can't find anything on the web to confirm this. Anyone got data?
Everyone asumes that all admins get mega bugs and sit in chilled rooms
ever admin I know Including me, Is over worked, Very under paid.
And the US is pretty much a labor market meaning
they guy down the hall that breaks down boxes
makes more than me.
If you'd look outside your immediate area, you'd find all kinds of positions. It is not tough for a sysadmin to find a comfortable living, so long as you're not determined to stay in Silicon Valley or similar.
Picking a day every other week to do a code review does wonders.
Employers look at this and say "What!? You want to dedicate 10% of our programmers' time to looking at code?" But they miss the point.
With code reviews, people become accountable for their code, and it quickly becomes apparent who is and isn't producing. This motivates people who don't want high visibility as slackers, and makes it easier to prove a case when you approach management about replacing somebody.
Let's see a sysadmin do the work of a construction worker at the construction worker's wage and see just how appreciated the sysadmin feels when he returns to his air conditioned cave with net access and computer toys and a paycheck three to four times as large as the manual laborer.
Lets revive the microwave beam weapons while we are at it. We'll pretend they are for disrupting electronics or radar mapping, but they also do a great job of interfering with brain activity. (You only have to head the brain a couple of degrees.) We'll make protective headgear for our soldiers.
I'm sorry - no matter how you present it, I still maintain the right to laugh at your tinfoil hat.
Celebrate the holiday by hosting an intervention. Take your favorite sysadmin out to TGIF or similar, and muster up the courage to say: "We're here to help ween you from User Friendly. It's just not funny. It's Ziggy, only with a narrower world view."
Every time you read a User Friendly strip, you should be reminded of the goofy daleks wheeling about and chanting "weeee are the superior beeeeings" and threatening everything with the only one tool they have. Tip one over and it's done for.
Taken to its extreme, the single ongoing punchline is: "Ha ha the people who actually do things with computers instead of fetish-fixating on the computers themselves are stupid! Ha ha we control the computers! Ha ha the people who actually do things are stupid again! Ha ha! We still control the computers!"
Being a sysadmin is cool and all. But generally speaking, you're one of the less valuable cogs in the machine which people are more eager to replace if you maintain that mindset.
Firstly, you can just boot off a floppy. The BIOS doesn't need to understand the drive, so long as the kernel does.
Secondly, you can likely boot off the drive if you go into the BIOS configuration and manually specify the correct number of heads and cylinders/sectors and just fib about the number of tracks. If you boot Linux from a small boot partition at the start of the drive, you're likely just fine.
I've had tremendous problems with media any time I burn over 12x. I've tried half a dozen brands which were rated for 24x or better, and even at 16x, they all have errors on about one disc in ten.
Do you have an IDE or SCSI drive?
I have seen IDE HDD -> SCSI CDRW always work perfectly, SCSI CDROM -> SCSI CDRW always work perfectly but IDE CDROM (48x) -> SCSI CDRW often fail and anything with IDE CDRW often fail.
I'm just curious if your problem is in fact the media or whether you are seeing increasing reliability problems as burn speed increases with an IDE CDRW drive. I'd like to know because I have an 8x SCSI CDRW drive and I'm about to buy a new CDRW drive for a friend. If the culprit is IDE I will buy him a 24x SCSI drive, if it really is the media I will buy him a 12x SCSI drive.
It's IDE, however I watch and the buffer never falls below 95% full for the duration of the burn, so it's never being starved.
I do also have a SCSI burner, which has the same problems at its max burn speed (16x).
I've had tremendous problems with media any time I burn over 12x. I've tried half a dozen brands which were rated for 24x or better, and even at 16x, they all have errors on about one disc in ten.
What kind of media do you use if you're shooting up to a 48x burn speed?
Someone else mentioned Cygwin, but I thought I'd provide more data.
You can download cygwin for free from cygwin.com. It includes both the client and the server for ssh. You can set up ssh as a service that runs even prior to login, so it's the real deal. All drives are accessible through the shell via the invisible/cygdrive/c,/cygdrive/d, etc directory. All the rest is explained on the Cygwin site. I believe commercial support for Cygwin is offered by Redhat, but it's worth noting that they have a very responsive free support list, frequented by all the major developers/porters.
If you want to read about how to keep the Debian packages that matter completely up to date, jump here.
Please point the next person to complain about Debian's slow releases here. The point remains: slow releases are still a good thing. For anyone with basic UNIX skills, the major updates are just a minor convenience. And each full distribution upgrade carries unnecessary risks.
There's little substitute for having models visible in your game engine. Even if the modeler matched the capabilities of your game, you're likely to have slight differences in your shading and lighting models, in your camera rules, and so on.
It's not that tough to attach a real-time exporter to most 3D modeling packages. Most of the major game development houses have some facility for displaying what's in a 3DS/Maya scene in a linked up game, whether it's running on an XBox, a GameCube or under Windows.
On top of that, you pretty much need to write your own tools to help artists spot abuse. Add a wireframe mode to your game, and tag all double-sided or currently-reversed polygons with a dot in the middle. Something that simple can point out poly overpopulation or wasted polys pretty quickly.
Also worth mention: The 17" iMac uses Geforce4MX. Not surprising, given that the Geforce2MX chipset isn't going to drive a digital display any larger than 1280x960, but still another worthy improvement.
Shame it's MX and not straight Geforce 4, but I keep forgetting that the iMac is supposed to be the "low end" machine. Because it sure doesn't look it, ya know?
Am I going to hell if I buy one just for movie watching and backup mail/dns on the home DSL?
Just one little interesting tidbit i noticed that is getting kind of lost in the noise: Did anyone else notice the little note on the Jaguar [apple.com] page? Apparently the 10.2 developer tools use GCC 3.1!
I found this interesting, as i had heard that the bulk of the linux distributions had not yet managed to migrate to GCC 3. Neat to see Apple is staying on top of this whole UNIX-technology thing:)
Is this really the case? It was my understanding that Linux distributions were holding back from a complete transition owing to 3.1 still being a bit buggy, or compiling kernels incorrectly.
Or perhaps Apple has some extra benefit to testing it can do, owing to only having to worry about the G4 processor?
Can you get your hands on an "expired" disc to look at what's on it?
Chances are it's just got a date or an ID on it, no signing or anything fancy like that, meant to keep stores from playing the wrong disc or from having playable music if they don't continue to subscribe.
I'd dump an image and look for something nonstandard in the TOC. If I were making a player that locked users out, I'd put it right there so I could use a standard CD player and just add code to compare the tail of that buffer to a 16 bit date number or such.
If you're really unlucky, they might actually be going so far as to put this on a special kind of limited use disc (a unique Disc Application Code in the Wobble Track), but it's unlikely they'd go to that expense unless this is a very popular and expensive service (which it may well be). At the least, I wouldn't be surprised if it were an audio disc and not a data CDR. I believe gcombust can tell you what DAC was read when a disc was inserted, and that might tell you more.
By the way, if it does have a special DAC, you're screwed without getting special media pressed or modifying the player. You can't write a wobble with a regular CD burner.
Aesthetically, that's a good way to make a small screen look even smaller. Cinching the sides like that draws a perceptual barrier between the screen and the writing area, making the thing look unusably small, even if the profile hasn't changed.
This might swell well in Japan where smaller is seen as better, but it seems more likely to stay on the shelves in the US for average consumers.
Guys, the PowerPoint slides for the Lucovsky presentation has been publicly downloadable for almost 2 years. I always find it sad when Slashdot reports something old as something new.
It was still probably news to most here. And it's interesting. Both make it a good story.
I dunno. Why don't you post your e-mail to slashdot, and we can find out...:)
I'll take you up on that one... I've setup uce@in-site.com.au to see how much spam a slashdot post can generate. I'll keep you updated...
That's not likely as effective as putting it in your user profile so it appears with every message you write.
For example, I've got my email address above, and I get well over a hundred pieces of spam a day. SpamAssassin tackles most of those, but I still have to tackle several per day.
No matter what Apple finally announces, at this or any other keynote, about half the comments in these forums will be filled with "no big deal" or "The Man is trying to crush us!" or some other wacky thing. Sometimes intelligent conversation about the Macintosh platform is hard to come by on slashdot. And that's a shame.
Okay, I'll bite. Maybe I'm feeding a troll, but --
Are you saying that "intelligent must equal enthusiastic" where Mac announcements are concerned? Or since you explicitly mention keynotes, do you only mean that anyone not excited by getting another stevejob is beneath contempt?
Slashdot Mirror
Most of these are extensions already existing for Windows and other OpenGL ports. NVidia example. ATI example. What they're basically saying is that the Mac drivers are caught up, and/or use of the extension is new to Jaguar's version of the Quartz engine.
There are a few Apple-specific extensions in there, but they're very specially purposed to Quartz' preferred data formats. Essentially, they're just a way to reduce the portability of the system (restricted pixel formats) in favor of some speed boosts, which is a pretty fair tradeoff if you're a company like Apple who only deals with a pocketful of vendors who make special concessions. You wouldn't want these back in OpenGL main.
There's some damned fine engineering going on at Apple, as always. But there's also the familiar nice spin, though. I wish they'd keep that much out of the technical presentations, or at least would more clearly mark it as such.
A previous employer told me that Shanghai was written by a disabled programmer, using Blow. I can't find anything on the web to confirm this. Anyone got data?
They're holding the floor down, taking a break from strenuous activity, whereas you're taking a break from clicking about to post on Slashdot?
If you'd look outside your immediate area, you'd find all kinds of positions. It is not tough for a sysadmin to find a comfortable living, so long as you're not determined to stay in Silicon Valley or similar.
Employers look at this and say "What!? You want to dedicate 10% of our programmers' time to looking at code?" But they miss the point.
With code reviews, people become accountable for their code, and it quickly becomes apparent who is and isn't producing. This motivates people who don't want high visibility as slackers, and makes it easier to prove a case when you approach management about replacing somebody.
Let's see a sysadmin do the work of a construction worker at the construction worker's wage and see just how appreciated the sysadmin feels when he returns to his air conditioned cave with net access and computer toys and a paycheck three to four times as large as the manual laborer.
Every time you read a User Friendly strip, you should be reminded of the goofy daleks wheeling about and chanting "weeee are the superior beeeeings" and threatening everything with the only one tool they have. Tip one over and it's done for.
Taken to its extreme, the single ongoing punchline is: "Ha ha the people who actually do things with computers instead of fetish-fixating on the computers themselves are stupid! Ha ha we control the computers! Ha ha the people who actually do things are stupid again! Ha ha! We still control the computers!"
Being a sysadmin is cool and all. But generally speaking, you're one of the less valuable cogs in the machine which people are more eager to replace if you maintain that mindset.
Maybe obvious. Just being anal.
You are so out of the ballpark. Buying a congressman for $500k is about as likely as buying a Cadillac with gumwrappers.
Secondly, you can likely boot off the drive if you go into the BIOS configuration and manually specify the correct number of heads and cylinders/sectors and just fib about the number of tracks. If you boot Linux from a small boot partition at the start of the drive, you're likely just fine.
I do also have a SCSI burner, which has the same problems at its max burn speed (16x).
What kind of media do you use if you're shooting up to a 48x burn speed?
You can download cygwin for free from cygwin.com. It includes both the client and the server for ssh. You can set up ssh as a service that runs even prior to login, so it's the real deal. All drives are accessible through the shell via the invisible /cygdrive/c, /cygdrive/d, etc directory. All the rest is explained on the Cygwin site. I believe commercial support for Cygwin is offered by Redhat, but it's worth noting that they have a very responsive free support list, frequented by all the major developers/porters.
Give it a go. I think you'll be impressed.
Please point the next person to complain about Debian's slow releases here. The point remains: slow releases are still a good thing. For anyone with basic UNIX skills, the major updates are just a minor convenience. And each full distribution upgrade carries unnecessary risks.
It's not that tough to attach a real-time exporter to most 3D modeling packages. Most of the major game development houses have some facility for displaying what's in a 3DS/Maya scene in a linked up game, whether it's running on an XBox, a GameCube or under Windows.
On top of that, you pretty much need to write your own tools to help artists spot abuse. Add a wireframe mode to your game, and tag all double-sided or currently-reversed polygons with a dot in the middle. Something that simple can point out poly overpopulation or wasted polys pretty quickly.
Shame it's MX and not straight Geforce 4, but I keep forgetting that the iMac is supposed to be the "low end" machine. Because it sure doesn't look it, ya know?
Am I going to hell if I buy one just for movie watching and backup mail/dns on the home DSL?
Not only is the display 17", at 1440 x 900 with square pixels, it's wide aspect. This is far better for movie playback.
I'm seriously tempted to get one just for watching DVDs.
Or perhaps Apple has some extra benefit to testing it can do, owing to only having to worry about the G4 processor?
Chances are it's just got a date or an ID on it, no signing or anything fancy like that, meant to keep stores from playing the wrong disc or from having playable music if they don't continue to subscribe.
I'd dump an image and look for something nonstandard in the TOC. If I were making a player that locked users out, I'd put it right there so I could use a standard CD player and just add code to compare the tail of that buffer to a 16 bit date number or such.
If you're really unlucky, they might actually be going so far as to put this on a special kind of limited use disc (a unique Disc Application Code in the Wobble Track), but it's unlikely they'd go to that expense unless this is a very popular and expensive service (which it may well be). At the least, I wouldn't be surprised if it were an audio disc and not a data CDR. I believe gcombust can tell you what DAC was read when a disc was inserted, and that might tell you more.
By the way, if it does have a special DAC, you're screwed without getting special media pressed or modifying the player. You can't write a wobble with a regular CD burner.
How do you know if you're at the cap?
This might swell well in Japan where smaller is seen as better, but it seems more likely to stay on the shelves in the US for average consumers.
For example, I've got my email address above, and I get well over a hundred pieces of spam a day. SpamAssassin tackles most of those, but I still have to tackle several per day.
Are you saying that "intelligent must equal enthusiastic" where Mac announcements are concerned? Or since you explicitly mention keynotes, do you only mean that anyone not excited by getting another stevejob is beneath contempt?