GIDs are stored in a list of 32-gid blocks, with an upper limit of 64k (posted recently on the Linux Kernel list).
See also, include/linux/sched.h, specifically the task_struct and group_info structs. (at least with 2.6.9)
Re:Gold? Ok... but tell me about a Linux client
on
Half Life 2 Goes Gold
·
· Score: 1
The Doom 3 client for Linux is out "already".
Indeedy. I plan on buying it RSN.
Re:Gold? Ok... but tell me about a Linux client
on
Half Life 2 Goes Gold
·
· Score: 1
HL1 is still 30$+ at retail stores
Sure... HL1 Platinum. HL1 + Many add-ons/expansions. I'd likely throw down $30 for that if I think I'll play it much.
I bought Max Payne 1 for $10 about 6mo or something (1year? Nah, I don't think it was that long) after it was released.
Re:Gold? Ok... but tell me about a Linux client
on
Half Life 2 Goes Gold
·
· Score: 1
Because until they get that second part down, the first is rather meaningless (for me at least).
Amen to that. HL1 was entertaining; I'll likely pick this up when it hits the $10 rack. No support for linux == I'm not buying it anytime soon.
They probably won't miss the 2% linux users out there, but we gotta vote with our wallet if we ever want to be heard. Same reason I need to get me Doom 3--Buy it when they release a Linux client, or get it Cheap if they don't. Show them we'll pay, but we won't be jerked around either.
And don't tell me to get Windows if I want to game.
I hear that. No reason to choose a platform for games--games simply aren't that important to me. For others games or a pet set of applications may well be that important. I try to vote when and where I can. Capitalism requires informed consumers who vote with their wallets to effect change.
Bottom line: I'll support you as well as you support me.:) I try to keep up with vendors if they give me good support (Yay Intel! Keep on giving us great drivers like you are, and you can be sure I'll buy your chips!) I'll very likely also buy the next release of Unreal Tournament for that reason as well.
My Zaurus gets used ten times more than my iPAQ. 9 hours battery life
Indeed. You may have a later revision than the parent did. Your experience echoes mine (with standard PDA use, I can go for one or two weeks without a recharge; I get about 8h if I'm playing mp3s with the screen off).
That said, that's all with the 3.10 ROM (I have an SL-5500). The 2.28 or whatever it was ROM left the wifi card on all the time, and was also just not nearly as power-conscious as the 3.10 ROM (which is standard in the SL-5600 and later, iirc). I used to have to recharge every night or so.
they have Opera? Since when?
They've had Opera since at least the 2.28 ROM, as I got it pre-installed on mine. The 3.10 ROM also has Opera. A much faster, more usable version, even. I would highly recommend the version of Opera that's on the 3.10 rom (6.something, iirc) for the Zaurus, as it has nifty features (like the ability to strip out some of the extraneous info and present a page in a PDA screen-compatible column instead of the PC screen-centric stuff that's standard fare nowadays.
But saying Microsoft caused BeOS's demise or that it was better than Windows is a bit of a stretch.
See, the problem is that, in the presence of a sinlge dominant competitor in the market (90+% is definitely that), you have to play a perfect game in order to merely survive. (And that's if luck's with you!)
Apple had its chance and blew it and is relegated to 2%. BeOS had its chance and blew it and died. MSFT has blown many things (like Windows ME!), yet they continue to survive. Remember the Microsoft Version 1.0 problem? They miss for the first couple of releases and then they succeed. Overwhelming market dominance (dare I say monopoly?) allows you to make mistakes without dying or becoming more or less irrelevant.
Which isn't to say that a monopolist can make as many mistakes as they want; IBM would be the poster child for that. But they have much, much greater margin for error than anyone who wishes to compete.
you could hook up a mythtv frontend on a mini-itx box with a flat-screen so that you can easily watch movies and tv and stuff while you're in the tub. Yay!
Actually, this is likely how I'm going to wire my home. One beefy central mythtv box with however many frontends scattered throughout the house.
Sure. My point was, though, that in Java, you don't add different components with the layout you want; you select a layout manager. In gtk, you don't select the layout manager, you add in components.
What do they know about the coming decade that we don't?
They know that they have a monpoly and a disinterested userbase and thus they know that they can push anything down our throats that they want and most people won't care.
We (anyone who's not part of Microsoft) have no such knowledge nor power.
Difference is that Java has different layout managers (GridBag being but one), whereas (so far as I can tell; am just a beginning gtk programmer), there is but one layout manager, but you can lay in components that effectively control the layout (hbox, vbox, etc.) In java, you select the layout manager; you don't put in different components that control layout.
Well I've got Cedega right here, and I've never got it to run anything more than 80% as fast as Windows.
On the same hardware? [shrug] Some games work better than others.
NVidia's drivers don't work on ATI hardware!
Indeedy. What I meant was this: When I purchased this notebook, I had the option of either buying an ATI or one of 2-3 NVidia cards. The ATI card has been problematic for me and others, while the NVidia card (aside from issues with suspending to disk/ram) works like a charm. That is what I meant.
Irrelevant. They're not high-performance chips, so driver optimization doesn't matter.
Not irrelevant to me (and, judging by the XOrg list, others as well). Unlike my ATI card, they work and have the best Linux support. This translates to, "While I may lose performance, I can actually use the 3d functionality!" Not irrelevant at all, chief.
How that's a fault of the modem (or me) is beyond me.
Order of at-faultness for this project not working on Linux, from maximum to minimum:
The modem vendor: For not supporting Linux
You: for not looking at what modems are and are not supported by Linux
Linux: if the devs don't have the specs, they can't write the driver. Not Linux's fault that nobody can get the specs or the vendor can't be bothered to write a driver.
The modem. It's just a piece of hardware and can't help being crappy.:)
If you choose step 2, you should see if your ideology isn't clouding your judgement. Just a thought.
There was a possible third option, namely Buy hardware supported by Linux or which supports Linux and then get it working in one afternoon (There are plenty of internal modems and HCF/HSF modems that work just fine under Linux). You'd also save yourself $200 or more (depending on what version of Windows you plan to use, + CALs).
If, on the other hand, you fully planned on using Windows from the beginning, then that's your call. But blaming Linux for not supporting hardware despite the fact that drivers for other hardware exist is faulty.
Also, the poster you're replying to needs to realize that yes, external modems are almost guaranteed to work with Linux (and are likely the best option imho), but that they aren't a panacaea, and won't work in every situation. In those cases, one might well look at the internal modems (preferably hardware, not HCF/HSF)).
Ehm, while true, there is a pool of developers that would work on a card were the specs available. I fall into that category, for one, along with others. Some new devs would also join, were they able to get the specs to their vid card. In addition, the current devs would be more productive and deliver better, more thorough drivers had they access to complete specs.
Now, if you say that specs are generally obtuse junk, and that you'd be better off not reading them due to their lousiness, then you may well have an argument.;)
Sure, having "the specs" would let you write a driver that's feature complete, but they won't help you make it fast.
Here, "feature complete" means "power management interface" and "3d acceleration". Most everything else is generally known by the devs (is why you can get 2d working under Linux with few problems generally, while you have to have the nvidia/ati binary drivers to actually have usable 3d.
A smart plan, because the only way Cedega will give you reasonable performance is to play a game that came out 18 months before you bought your PC.
Incorrect. According to acquaintances I trust, cedega (or was it just wine?) runs some games faster than Windows. But a native Linux version should generally run faster (given the same time investment/knowledge/love as the Windows port).
The basic point of what I was saying was that, if ATI can't make drivers that people can use, they should give others a chance to step up to the plate. NVidia's drivers have no problems on this same hardware, from what I can tell (from looking at the Dell forums). Apparently Intel and VIA have released Free drivers, and are thus now either in the kernel or in DRI's CVS. If I'd have chosen either of those (plentiful) routes instead of the ATI one, I'd be setting pretty. Guess I'm just a dissatisfied customer. I just hope that they will back up the (only) reason I bought ATI: they give devleopers documentation a little while down the road.
I was quite impressed by UT2K4 on Linux. It was rip-roarin' fast, even at 1920x1200!. Well, until those danged ATI drivers crapped out (they always do, after running for about 3 minutes, they get all sorts of horizontal lines chopping up the view and the game becomes unplayable. Indeed, no 3D works from that point on (will get the horizontal lines after a few tens of seconds, length of time is reduced every time I start playing again without rebooting).
I was also impressed with the Castle Wolfenstein Linux game as well as Neverwinter Nights. If only I had gotten an NVidia (or ATI made drivers that worked for me, or they released the specs or....)
Generally, I will try to buy a game soon after they release the Linux client (and when budget permits). I may buy a new windows-only game if I hear it was really good and it's on massive discount (e.g. the $10 rack).
Why? Does Linux have some sort of self-imposed deadline for world domination?
Yes. Whenever Microsoft's Palladium (or equivalent from another vendor) gets shoved into all PCs by federal (US, EU, et al.) "intellectual property" regulations. I.e. 2011.
Quite possibly. I don't actually do any reverse-enginnering of the file system. However, my point was that such reverse-engineering, particularly when it has to be 100% or (potentially vital!) data is lost, is a very time-intensive process.
I suspect it has something to do with having to boot into Windows, change something, shut down windows, and then look at a huge disk image to see what changed.
Specs would save a ton of dev work. Do you hear me, DoJ?
Slashdot Mirror
Please mod parent up.
GIDs are stored in a list of 32-gid blocks, with an upper limit of 64k (posted recently on the Linux Kernel list).
See also, include/linux/sched.h, specifically the task_struct and group_info structs. (at least with 2.6.9)
Indeedy. I plan on buying it RSN.
Sure... HL1 Platinum. HL1 + Many add-ons/expansions. I'd likely throw down $30 for that if I think I'll play it much.
I bought Max Payne 1 for $10 about 6mo or something (1year? Nah, I don't think it was that long) after it was released.
Amen to that. HL1 was entertaining; I'll likely pick this up when it hits the $10 rack. No support for linux == I'm not buying it anytime soon.
They probably won't miss the 2% linux users out there, but we gotta vote with our wallet if we ever want to be heard. Same reason I need to get me Doom 3--Buy it when they release a Linux client, or get it Cheap if they don't. Show them we'll pay, but we won't be jerked around either.
I hear that. No reason to choose a platform for games--games simply aren't that important to me. For others games or a pet set of applications may well be that important. I try to vote when and where I can. Capitalism requires informed consumers who vote with their wallets to effect change.
Bottom line: I'll support you as well as you support me. :) I try to keep up with vendors if they give me good support (Yay Intel! Keep on giving us great drivers like you are, and you can be sure I'll buy your chips!) I'll very likely also buy the next release of Unreal Tournament for that reason as well.
Indeed. You may have a later revision than the parent did. Your experience echoes mine (with standard PDA use, I can go for one or two weeks without a recharge; I get about 8h if I'm playing mp3s with the screen off).
That said, that's all with the 3.10 ROM (I have an SL-5500). The 2.28 or whatever it was ROM left the wifi card on all the time, and was also just not nearly as power-conscious as the 3.10 ROM (which is standard in the SL-5600 and later, iirc). I used to have to recharge every night or so.
They've had Opera since at least the 2.28 ROM, as I got it pre-installed on mine. The 3.10 ROM also has Opera. A much faster, more usable version, even. I would highly recommend the version of Opera that's on the 3.10 rom (6.something, iirc) for the Zaurus, as it has nifty features (like the ability to strip out some of the extraneous info and present a page in a PDA screen-compatible column instead of the PC screen-centric stuff that's standard fare nowadays.
See, the problem is that, in the presence of a sinlge dominant competitor in the market (90+% is definitely that), you have to play a perfect game in order to merely survive . (And that's if luck's with you!)
Apple had its chance and blew it and is relegated to 2%. BeOS had its chance and blew it and died. MSFT has blown many things (like Windows ME!), yet they continue to survive. Remember the Microsoft Version 1.0 problem? They miss for the first couple of releases and then they succeed. Overwhelming market dominance (dare I say monopoly?) allows you to make mistakes without dying or becoming more or less irrelevant.
Which isn't to say that a monopolist can make as many mistakes as they want; IBM would be the poster child for that. But they have much, much greater margin for error than anyone who wishes to compete.
you could hook up a mythtv frontend on a mini-itx box with a flat-screen so that you can easily watch movies and tv and stuff while you're in the tub. Yay!
Actually, this is likely how I'm going to wire my home. One beefy central mythtv box with however many frontends scattered throughout the house.
that there is no Linux or Mac gaming market!
I mean, look at it, they have Windows, Windows, and more Windows! Obviously, only Windows users buy games!
[end sarcasm]
And claiming compliance with Windows (i.e. the logo; same as with LSB) costs you what? Anybody?
Bueller?
Right. I dunno if I like gtk or java's more; they have their high points and their low points. *shrug*
Sure. My point was, though, that in Java, you don't add different components with the layout you want; you select a layout manager. In gtk, you don't select the layout manager, you add in components.
Small difference in some ways, big in others.
They know that they have a monpoly and a disinterested userbase and thus they know that they can push anything down our throats that they want and most people won't care.
We (anyone who's not part of Microsoft) have no such knowledge nor power.
They're certainly similar.
Difference is that Java has different layout managers (GridBag being but one), whereas (so far as I can tell; am just a beginning gtk programmer), there is but one layout manager, but you can lay in components that effectively control the layout (hbox, vbox, etc.) In java, you select the layout manager; you don't put in different components that control layout.
Additionally, for more data, apps can use shared memory locally.
On the same hardware? [shrug] Some games work better than others.
Indeedy. What I meant was this: When I purchased this notebook, I had the option of either buying an ATI or one of 2-3 NVidia cards. The ATI card has been problematic for me and others, while the NVidia card (aside from issues with suspending to disk/ram) works like a charm. That is what I meant.
Not irrelevant to me (and, judging by the XOrg list, others as well). Unlike my ATI card, they work and have the best Linux support. This translates to, "While I may lose performance, I can actually use the 3d functionality!" Not irrelevant at all, chief.
Order of at-faultness for this project not working on Linux, from maximum to minimum:
There was a possible third option, namely Buy hardware supported by Linux or which supports Linux and then get it working in one afternoon (There are plenty of internal modems and HCF/HSF modems that work just fine under Linux). You'd also save yourself $200 or more (depending on what version of Windows you plan to use, + CALs).
If, on the other hand, you fully planned on using Windows from the beginning, then that's your call. But blaming Linux for not supporting hardware despite the fact that drivers for other hardware exist is faulty.
Also, the poster you're replying to needs to realize that yes, external modems are almost guaranteed to work with Linux (and are likely the best option imho), but that they aren't a panacaea, and won't work in every situation. In those cases, one might well look at the internal modems (preferably hardware, not HCF/HSF)).
Ehm, while true, there is a pool of developers that would work on a card were the specs available. I fall into that category, for one, along with others. Some new devs would also join, were they able to get the specs to their vid card. In addition, the current devs would be more productive and deliver better, more thorough drivers had they access to complete specs.
Now, if you say that specs are generally obtuse junk, and that you'd be better off not reading them due to their lousiness, then you may well have an argument.
Here, "feature complete" means "power management interface" and "3d acceleration". Most everything else is generally known by the devs (is why you can get 2d working under Linux with few problems generally, while you have to have the nvidia/ati binary drivers to actually have usable 3d.
Incorrect. According to acquaintances I trust, cedega (or was it just wine?) runs some games faster than Windows. But a native Linux version should generally run faster (given the same time investment/knowledge/love as the Windows port).
The basic point of what I was saying was that, if ATI can't make drivers that people can use, they should give others a chance to step up to the plate. NVidia's drivers have no problems on this same hardware, from what I can tell (from looking at the Dell forums). Apparently Intel and VIA have released Free drivers, and are thus now either in the kernel or in DRI's CVS. If I'd have chosen either of those (plentiful) routes instead of the ATI one, I'd be setting pretty. Guess I'm just a dissatisfied customer. I just hope that they will back up the (only) reason I bought ATI: they give devleopers documentation a little while down the road.
I was quite impressed by UT2K4 on Linux. It was rip-roarin' fast, even at 1920x1200!. Well, until those danged ATI drivers crapped out (they always do, after running for about 3 minutes, they get all sorts of horizontal lines chopping up the view and the game becomes unplayable. Indeed, no 3D works from that point on (will get the horizontal lines after a few tens of seconds, length of time is reduced every time I start playing again without rebooting).
I was also impressed with the Castle Wolfenstein Linux game as well as Neverwinter Nights. If only I had gotten an NVidia (or ATI made drivers that worked for me, or they released the specs or....)
Generally, I will try to buy a game soon after they release the Linux client (and when budget permits). I may buy a new windows-only game if I hear it was really good and it's on massive discount (e.g. the $10 rack).
Actually, D-Bus is merely a message-passing system. HAL builds on it to do hardware notifications, IIUC (u = understand).
You can make D-BUS enabled apps that have nothing to do with HAL or G-V-M.
The GIMP can take screen or window shots. I suspect that some WMs/DEs will take a screenshot with PrintScrn. Yup. GNOME 2.6 does at least. :)
Yes. I've not personally done it, but there is documentation on it on the mythtv site.
You can have multiple backends and multiple frontends, on different machines. There should be guides for getting this going.
Honestly, I don't know wny schools don't do this, what with a nice PC in every classroom, and projectors being fairly common (at least in the US).
Indeed. That was the first thing that came to my mind as well!
Seriously, MythTV is what you are looking for, chief.
Yes. Whenever Microsoft's Palladium (or equivalent from another vendor) gets shoved into all PCs by federal (US, EU, et al.) "intellectual property" regulations. I.e. 2011.
Quite possibly. I don't actually do any reverse-enginnering of the file system. However, my point was that such reverse-engineering, particularly when it has to be 100% or (potentially vital!) data is lost, is a very time-intensive process.
I suspect it has something to do with having to boot into Windows, change something, shut down windows, and then look at a huge disk image to see what changed.
Specs would save a ton of dev work. Do you hear me, DoJ?