He chose some pretty bad examples of bad error handling - they all gave the module and direct cause of the error, and provided ample clues for the programmer in each case to go find out what went wrong. If we're looking for anything that open source programmers do that stinks, it's making GUI apps that pretend nothing's wrong.
Way to go, I say. Would rather have hugely detailed warnings any day.
So intel's plan to bolster their flagging market share is to introduce an entirely new platform that's not backwards compatible?
Yup, and it would have worked too (if it wasn't for you pesky kids) had the chip come out when it was supposed to. Two, maybe three years ago, with the current level of performance.
Part of what pisses me off about this whole IA-64 thing is that it was actually quite a good idea.
The RealTek 8139 PCI NIC redefines the meaning of 'low end.'
This can only be because he hasn't had much experience with the Davicom chipsets. I had a problem with the Davicom chips dropping off at layer 2 under load. Posts to freebsd-hackers seem to suggest that the answer is "yeah, they do don't they" - so bizarrely this is an area where realtek's have an advantage.
I'm no fan of 3com either. I have a 3c905b in a FreeBSD box that (for one reason or another) I just don't want to have to dig out and change. Recently we have been having difficulties running the chip at 100MBit/sec, both on hubs and switches. Changed the cable, even shielded it to no avail.
Where does this leave us? Intel 82559 or death! Although these also have a disclaimer about not wanting to be connected via loopback cables. Gnnnn!
Yes and no. Sure, I never used pipes on windows but I did use the message passing mechanism some times. Interestingly enough this is the low level mechanism used by COM to do interprocess comms.
That would have been a better test.
BTW, The source code examples in the article did a great job of reminding me why I hate coding for windows:)
Apple created this, let them have it. Either come up with something better or stop snickering about Microsoft and 'innovation'.
Never a truer word. And make it faster, for gods' sake. I'm sitting here using KDE on a P3-933 with 128Mb and it's a dog. Fast enough, sure, but still a dog.
Sometimes I wonder how programmers worked before widespread use of the internet.
Books. MSDN (believe it or not).
I can sometimes loose chunks of time posting stuff to slashdot, surfing the web etc that is not really work related.
Don't worry, before the Internet there was staring out the window, drinking too much coffee... hundreds of ways of wasting time. On the flip side, if you feel your productivity is suffering because you're posting to slashdot, don't. Basically. All the firewalling, content filters and Nazi'ism from sysadmin's will not stop bored people from wasting time.
Not the end of the world for Exodus I mean. Obviously it isn't the end of the world for everyone else.
IIRC Exodus have fourty something hosting facilities, presumably mostly empty right now, but they do have a lot of work and an excellent reputation. They'll be fine, they just need to downscale by a factor of, like, five. This, and the refinancing by GE, are just a part of that process.
OK, so a week ago we have the FBI complaining that/bin/laden has given up on his satellite phones and computers in favour of sending messengers on camels (or whatever, apologies for lame western stereotyping) since they can't be wiretapped. Now: Please r00t his servers for us.
Piss poor. And anyway, a whole bunch of people are going to want a whole load of legal protection for what they know before they will do this. I mean, spend the next six months breaking terrorists servers apart and in a years time when.Net takes it up the arse... well, they're going to know whose door to break down, aren't they?
But if anyone goes for this: good luck. I do have some pretty serious problems with piling into Afghanistan and blowing shit up; but no moral qualms whatsoever when it comes to merely trashing terrorists ability to operate.
the porting effort to move code the likes of PHP/JSP/servelets
...is all well and good, but most people using IIS are doing so because they are using ASP. Now, while you can get a relatively inexpensive ASP engine from Sun (via Chilisoft) it won't allow you to use any custom ActiveX's... and I've not heard any news on ADO compatibility or similar japes.
Anyone got any experience with Chilisoft ASP? I'm considering using it for a project here and would like to know how transparent it is for the ASP weenies...
Using VNC? Oh yeah, that'll need broadband OK. I can see a pattern emerging here:
Java->slow language, made by people that sell big computers.
VNC netphones->bandwidth heavy, made by people that sell fat pipes.
Hmmm. I guess that whole better mousetrap thing is being forgotten again.
Dave
Re:what is it good for?
on
2.2 GHz Xeon
·
· Score: 3, Insightful
There is a big difference between 40 and 100 fps, particularly if you have a monitor that refreshes at 100Hz. Not wishing to harp on and get even more off topic, but if you turn 180 degrees, aim and fire in a quarter second that gives you:
40Hz 100Hz (apologies for lame fmt)
Turn 6 15
Aim 3 8
Fire 1 2
..frames in which to perform the operation. Those couple of extra frames for aiming with actually do make a difference. I actually got noticeably better at Quake (particularly the rail gun) when moving to a faster card and an optical mouse.
Back on topic, this is good for databases. Faster processors means fewer processors, less cache contention, good for all concerned really. And this is a really good move on Intel's part - rather than support a.18 micron part they really wish they hadn't made, in a socket they will use for precisely one processor, on a chipset for precisely one processor, they're swallowing pride and just getting on to the next one. Sensible. Very. Particularly when you consider how much the P4 needs its' on die cache to come up before it starts working properly.
What about Unreal 2 or do you not think that it will raise the bar on computer hardware needed.
Nah, not really. I'm currently running a 1GHz Atlon with a GF2 GTS. In other words, entirely middle of the road by geek standards:) If Unreal 2 needs bigger hardware than this, or this overclocked a bit, then it can get stuffed. I'm just not going to get that excited about Unreal2, it just looks like more polygons really.
Any word on a *nix or mac release of this, because i am really getting tired of being in windows for my games.
The doom engine will be portable, because John Carmack's a portability fanatic. The only public demonstration of Doom3 was on a Mac, after all. I wouldn't hold your breath on Unreal though.
I suppose they signed some agreement years ago and now they're stuck.
Yeah, that's exactly it. I believe they promised to only (ONLY!!) use Rambus and have been trying to wheedle their way out of it ever since. I believe they also bought a shitload of Rambus stock, then promised to only use Rambus, then sold the stock so it serves them right.
Will this be the next "big game" that will force users (er, gamers) to upgrade
No, leave that for Doom3 and Quake4 (both based on the new Doom engine). That is promising to make ALL current systems look death-defyingly shittly. And yes, that does include the Geforce3 owners of this world, but quite possibly not the next generation Radeon (that you can't buy yet).
What RTCW is going to do is make me feel less stupid for buying a 1GHz Athlon a couple of weeks back. I mean, 150fps in Q3, do I need it? No. Am I going to need it for RTCW? Quite possibly.
Right, I'm off home, guess what I started downloading before I came to work?
Hell yes, I've been saying that for ages. In fact, don't just let it apply to Microsoft, there should be a general "it's my data" law something along the lines of:
A third party performing a service in data storage, transmission or processing must, on demand, reveal details of exactly how that data is stored or transmitted.
This would be a good move because while it gives MS their right to, ahhh, innovate it also stops the other wankers in the industry (that's you, Sun) from trying to pull the same thing.
I'm also not sure that the API's are the crown jewels here either. As we have seen, it's the transmission and storage of data that make the difference - the Exchange wire protocol, butchered Kerberos protocol and Word file formats being the cases that come most immediately to mind. Strict limits on breaking compatibility are probably not necessary since MS got where they are today by ensuring backwards compatibility to an almost anally retentive degree.
Sack of shit. BeOS started its' life on PowerPC and was ported to x86 in a matter of weeks. _weeks_. Palm can port BeOS to whatever they want but you have to ask if there's a point. For instance - a lot of the stuff in BeOS was to do with SMP, is that really relevant in a handheld? Ditto a journaling filesystem or Posix compliance.
I havent heard much of the EPoc/Psion-base Symbian platform
Symbian are doing just fine. Maybe a bit behind where they ought to be, but the uber cool new Nokia communicators are out, available, can be bought. Symbian also have a more interesting view of OS design - they design it to be as light as possible, getting the processor back to its' sleep state quickly. While this doesn't matter a jot for a desktop OS, it's critical for those markets where reducing power consumption and hence battery size is important. Anything that goes in your pocket in other words.
At least we'll have plenty of doctors and lawyers who can code too
Maybe it was a hugely cunning way of making the professional population tech-savvy. OR: Maybe it was a hugely-HUGELY cunning way of making the professional population Microsoft haters.
Sadly there is a status symbol with software - or rather computers in general. Explain the existence of thin laptops, for instance.
Dave
Re:Implications for alpha?
on
HP Buys Compaq
·
· Score: 3, Insightful
Ooooohhh, sounds good! I know that StrongARM was reckoned to be the most Mips/Watt, is it still ahead when we get to the - kinda - near gig levels that we're talking about here?
Maybe I'm going to have to learn ARM assembly after all:)
Dave
Re:Implications for alpha?
on
HP Buys Compaq
·
· Score: 2
Power/PPC - Motorola has almost given up
On the contrary, I think Motorola are having a cunning reposition of the PPC processor to future embedded platforms. If you're running fanless it's between that and StrongARM - and that's showing no signs of going 500MHz+ in the near future.
Slashdot Mirror
He chose some pretty bad examples of bad error handling - they all gave the module and direct cause of the error, and provided ample clues for the programmer in each case to go find out what went wrong. If we're looking for anything that open source programmers do that stinks, it's making GUI apps that pretend nothing's wrong.
Way to go, I say. Would rather have hugely detailed warnings any day.
Dave
So intel's plan to bolster their flagging market share is to introduce an entirely new platform that's not backwards compatible?
Yup, and it would have worked too (if it wasn't for you pesky kids) had the chip come out when it was supposed to. Two, maybe three years ago, with the current level of performance.
Part of what pisses me off about this whole IA-64 thing is that it was actually quite a good idea.
Dave
I'm just wondering what Nautilus is *doing* in that 11 seconds it takes to load a folder.
IIRC Monsieur Cox was saying something about loading a font 4700 times. Hmmm.
Dave
The RealTek 8139 PCI NIC redefines the meaning of 'low end.'
This can only be because he hasn't had much experience with the Davicom chipsets. I had a problem with the Davicom chips dropping off at layer 2 under load. Posts to freebsd-hackers seem to suggest that the answer is "yeah, they do don't they" - so bizarrely this is an area where realtek's have an advantage.
I'm no fan of 3com either. I have a 3c905b in a FreeBSD box that (for one reason or another) I just don't want to have to dig out and change. Recently we have been having difficulties running the chip at 100MBit/sec, both on hubs and switches. Changed the cable, even shielded it to no avail.
Where does this leave us? Intel 82559 or death! Although these also have a disclaimer about not wanting to be connected via loopback cables. Gnnnn!
Dave
Yes and no. Sure, I never used pipes on windows but I did use the message passing mechanism some times. Interestingly enough this is the low level mechanism used by COM to do interprocess comms.
:)
That would have been a better test.
BTW, The source code examples in the article did a great job of reminding me why I hate coding for windows
Dave
Apple created this, let them have it. Either come up with something better or stop snickering about Microsoft and 'innovation'.
Never a truer word. And make it faster, for gods' sake. I'm sitting here using KDE on a P3-933 with 128Mb and it's a dog. Fast enough, sure, but still a dog.
Dave
Sometimes I wonder how programmers worked before widespread use of the internet.
Books. MSDN (believe it or not).
I can sometimes loose chunks of time posting stuff to slashdot, surfing the web etc that is not really work related.
Don't worry, before the Internet there was staring out the window, drinking too much coffee... hundreds of ways of wasting time. On the flip side, if you feel your productivity is suffering because you're posting to slashdot, don't. Basically. All the firewalling, content filters and Nazi'ism from sysadmin's will not stop bored people from wasting time.
Speaking of which, I have stuff to do.
Dave
Not the end of the world for Exodus I mean. Obviously it isn't the end of the world for everyone else.
IIRC Exodus have fourty something hosting facilities, presumably mostly empty right now, but they do have a lot of work and an excellent reputation. They'll be fine, they just need to downscale by a factor of, like, five. This, and the refinancing by GE, are just a part of that process.
Dave
OK, so a week ago we have the FBI complaining that /bin/laden has given up on his satellite phones and computers in favour of sending messengers on camels (or whatever, apologies for lame western stereotyping) since they can't be wiretapped. Now: Please r00t his servers for us.
.Net takes it up the arse... well, they're going to know whose door to break down, aren't they?
Piss poor. And anyway, a whole bunch of people are going to want a whole load of legal protection for what they know before they will do this. I mean, spend the next six months breaking terrorists servers apart and in a years time when
But if anyone goes for this: good luck. I do have some pretty serious problems with piling into Afghanistan and blowing shit up; but no moral qualms whatsoever when it comes to merely trashing terrorists ability to operate.
Dave
the porting effort to move code the likes of PHP/JSP/servelets
... and I've not heard any news on ADO compatibility or similar japes.
...is all well and good, but most people using IIS are doing so because they are using ASP. Now, while you can get a relatively inexpensive ASP engine from Sun (via Chilisoft) it won't allow you to use any custom ActiveX's
Anyone got any experience with Chilisoft ASP? I'm considering using it for a project here and would like to know how transparent it is for the ASP weenies...
Dave
Using VNC? Oh yeah, that'll need broadband OK. I can see a pattern emerging here:
Java->slow language, made by people that sell big computers.
VNC netphones->bandwidth heavy, made by people that sell fat pipes.
Hmmm. I guess that whole better mousetrap thing is being forgotten again.
Dave
There is a big difference between 40 and 100 fps, particularly if you have a monitor that refreshes at 100Hz. Not wishing to harp on and get even more off topic, but if you turn 180 degrees, aim and fire in a quarter second that gives you:
.18 micron part they really wish they hadn't made, in a socket they will use for precisely one processor, on a chipset for precisely one processor, they're swallowing pride and just getting on to the next one. Sensible. Very. Particularly when you consider how much the P4 needs its' on die cache to come up before it starts working properly.
40Hz 100Hz (apologies for lame fmt)
Turn 6 15
Aim 3 8
Fire 1 2
..frames in which to perform the operation. Those couple of extra frames for aiming with actually do make a difference. I actually got noticeably better at Quake (particularly the rail gun) when moving to a faster card and an optical mouse.
Back on topic, this is good for databases. Faster processors means fewer processors, less cache contention, good for all concerned really. And this is a really good move on Intel's part - rather than support a
Mind you, it'd cost the same as my house.
Dave
Where are moderation points when you need them?
:)
While I've got your ear Jordan, not that I have, I thought we might be looking at XFree4 in this release? Is the support still not really there?
Sorry to hear about the delays on SMPng. If I could help I would, but as you are aware this is kinda specialist work.
Hope you're enjoying Apple.
Dave
You run the danger of blocking a request that's coming via a transparent proxy. Blocking it would block everyone behind that proxy.
Comments?
Dave
An AMD Athlon should do you quite nicely :)
Dave
What about Unreal 2 or do you not think that it will raise the bar on computer hardware needed.
:) If Unreal 2 needs bigger hardware than this, or this overclocked a bit, then it can get stuffed. I'm just not going to get that excited about Unreal2, it just looks like more polygons really.
Nah, not really. I'm currently running a 1GHz Atlon with a GF2 GTS. In other words, entirely middle of the road by geek standards
Any word on a *nix or mac release of this, because i am really getting tired of being in windows for my games.
The doom engine will be portable, because John Carmack's a portability fanatic. The only public demonstration of Doom3 was on a Mac, after all. I wouldn't hold your breath on Unreal though.
Dave
I suppose they signed some agreement years ago and now they're stuck.
Yeah, that's exactly it. I believe they promised to only (ONLY!!) use Rambus and have been trying to wheedle their way out of it ever since. I believe they also bought a shitload of Rambus stock, then promised to only use Rambus, then sold the stock so it serves them right.
Athlon good. DDR gooder. Interlaced DDR (nVidia crush) goodest.
Dave
Will this be the next "big game" that will force users (er, gamers) to upgrade
:)
No, leave that for Doom3 and Quake4 (both based on the new Doom engine). That is promising to make ALL current systems look death-defyingly shittly. And yes, that does include the Geforce3 owners of this world, but quite possibly not the next generation Radeon (that you can't buy yet).
What RTCW is going to do is make me feel less stupid for buying a 1GHz Athlon a couple of weeks back. I mean, 150fps in Q3, do I need it? No. Am I going to need it for RTCW? Quite possibly.
Right, I'm off home, guess what I started downloading before I came to work?
Dave
Couldn't agree with you more. It's a good language, Microsoft invented it, nobody uses it. So bite me, no troll here.
Dave
Hell yes, I've been saying that for ages. In fact, don't just let it apply to Microsoft, there should be a general "it's my data" law something along the lines of:
A third party performing a service in data storage, transmission or processing must, on demand, reveal details of exactly how that data is stored or transmitted.
This would be a good move because while it gives MS their right to, ahhh, innovate it also stops the other wankers in the industry (that's you, Sun) from trying to pull the same thing.
I'm also not sure that the API's are the crown jewels here either. As we have seen, it's the transmission and storage of data that make the difference - the Exchange wire protocol, butchered Kerberos protocol and Word file formats being the cases that come most immediately to mind. Strict limits on breaking compatibility are probably not necessary since MS got where they are today by ensuring backwards compatibility to an almost anally retentive degree.
Anyway, work beckons.
Dave
BeOS runs on x86 only
Sack of shit. BeOS started its' life on PowerPC and was ported to x86 in a matter of weeks. _weeks_. Palm can port BeOS to whatever they want but you have to ask if there's a point. For instance - a lot of the stuff in BeOS was to do with SMP, is that really relevant in a handheld? Ditto a journaling filesystem or Posix compliance.
I havent heard much of the EPoc/Psion-base Symbian platform
Symbian are doing just fine. Maybe a bit behind where they ought to be, but the uber cool new Nokia communicators are out, available, can be bought. Symbian also have a more interesting view of OS design - they design it to be as light as possible, getting the processor back to its' sleep state quickly. While this doesn't matter a jot for a desktop OS, it's critical for those markets where reducing power consumption and hence battery size is important. Anything that goes in your pocket in other words.
Dave
At least we'll have plenty of doctors and lawyers who can code too
Maybe it was a hugely cunning way of making the professional population tech-savvy. OR: Maybe it was a hugely-HUGELY cunning way of making the professional population Microsoft haters.
Progress!
Dave
Sadly there is a status symbol with software - or rather computers in general. Explain the existence of thin laptops, for instance.
Dave
Ooooohhh, sounds good! I know that StrongARM was reckoned to be the most Mips/Watt, is it still ahead when we get to the - kinda - near gig levels that we're talking about here?
:)
Maybe I'm going to have to learn ARM assembly after all
Dave
Power/PPC - Motorola has almost given up
On the contrary, I think Motorola are having a cunning reposition of the PPC processor to future embedded platforms. If you're running fanless it's between that and StrongARM - and that's showing no signs of going 500MHz+ in the near future.
I think PPC has a bright future, actually.
Dave