Blame driver vendors for thinking desktop users don't care. And blame desktop users for generally not caring, or even knowing it's possible for that matter.
Right, these clusters of behaviors and conditions that we've identified and labelled and which you even admit to existing, have no meaning. It's all just part of life's rich tapestry, and that's that.
Similarly, there's no such thing as planets, stars, moons, etc; there's just clusters of matter with various characteristics and states that tend to occur together throughout the observed universe -- the labels we've applied to them are meaningless buzzwords, and in 100 years we'll have moved on to the next stupid faddy names.
And of course since I've never met Bram, and know next to nothing about him beyond one bit of software he wrote and an interview or two he did, I can feel confident in saying that his behavior isn't worthy of a label, even if it might help him and everyone else cope with and understand his problems. They certainly haven't impacted his life in such a profound way that being able to talk about them might hold any importance to him or any of the similarly geeky people likely to be reading his interviews.
What, is there something about certain Rails components that replacing one with something that better suits your requirements (and which is largely API compatible, I might add) will cause the magic smoke to escape?
Yes, Rails has limitations; so does everything else. That's why we design systems that are loosely coupled so components can be swapped out as and when needed. Ruby is especially good at this, and Rails is no exception.
Object Graphs is an alternative to ActiveRecord that has quite flexible support for mapping legacy schemas to your objects. Nitro itself looks like a nice alternative to Rails too, but you can mix and match if you feel like it.
I have no idea, really, I'm just speculating; I suspect it's not as simple as that. They're already hideously complicated and expensive to design; putting two on a chip, or even just two on a card, may well prove to be easier than working out how to integrate 700 million transistors into a single core. *shrug*
I started with the GeForce 256DDR (T&L, woo!), Ti4200, 5900XT, 6800GT and now a 7800GTX. I'm also okay with this:)
That's over £2k without storage, hardware RAID, graphics, etc, with a fairly modest amount of wriggle room for further savings. I hope you have a really good use for all those cores, or at least enough money that the cost is largely irrelevent. You save a lot going dual single core, and there are still reasons you might prefer a simple dual Opteron over an X2 (like support for >4G of memory and more readily available PCI-X), but if you're just looking at a nice desktop rather than a professional graphics workstation or big multi-user server, they probably don't apply much in this case.
Personally, I went for a mid-range X2, left the dual dual Opteron in the data centre and am awaiting a handmedown IBM eServer for my home serverish needs. YMMV.
Soft shadows in games like Chronices of Riddick and FEAR really take it out of my 7800GTX, especially at my TFT's native resolution (1600*1200). SLI's probably pretty much the only way to play at such high settings reasonably. Complex maps/situations in other games can also make it chug, and I'm sure it only gets worse at higher resolutions and AA levels (I normally play with 4X AA, an SLI user will probably be breezing along at 16X).
Then of course there's the people who use 3D hardware as part of their job; CAD, 3D artists, level designers, game engine developers; one of the first SLI forum threads I read was by a guy involved in medical imaging. SLI is also laying the groundwork for future multicore cards; in much the same way that SMP has been the realm of rich bastards and high end professional users until multicore consumer level CPU's, SLI will probably remain in the realm of the same sort of people for a year or two until we start seeing multicore NV chips.
Elite II (Frontier) had pretty realistic physics; I think it's a shame more games didn't try to run with that, and come up with novel/fun ways of dealing with the huge scales involved, instead of copping out with small environments and mysterious drag keeping everything within a few hundred m/s.
DESCRIPTION
The killall utility kills processes selected by name, as opposed to the
selection by pid as done by kill(1). By default, it will send a TERM
signal to all processes with a real UID identical to the caller of
killall that match the name procname. The super-user is allowed to kill
any process.
Though you need to be careful with that if you admin other systems like Solaris;)
pkill/grep are nice too, and are standard on a fair few systems now.
Well, presumably they've improved somewhat since the cheapy chipsets I played with last year then; they could hardly have gotten worse;)
At ~30MB/s I see something like 3% CPU use on my X2 4400+ and an on-board PCI-E Marvell; once I get my fileserver sorted out with PCI-X (surplus rackmount server) I'd hope to see more exciting transfer rates more in tune with modern HD's. Presumably similar could be said with two consumer-level PCI-E systems now; that makes me very happy:)
For comparison, on a 3200+ I see ~30% CPU use on an E1000 and ~80% on a Realtek using similar transfer rates, so you see where my hate for rl comes from;)
To hell with that; GigE can totally hammer the dinky PCI bus most systems have these days, you want something with proper interrupt moderation, 100% working checksum offload and decent DMA support; jumbo frames are nice too. Cheapy Realtek cards need not apply; spend a bit more on an Intel Pro/1000 or so, unless you like having 80% of your CPU dedicated to servicing interrupts and copying data around pointlessly.
PCI-E stuff seems much better, thankfully; makes on-board GigE on consumer level systems actually quite nice.
Yup, Aliens. The scene in the movie was quite weak so didn't make it to the theatrical release, but you can still see it (among other deleted scenes) in the Special Edition DVD (R2 at least).
I wrote my own ickle ad killer for Opera using CSS, which mostly scratches my itch. I'm sure there are better ones out there, but I haven't looked too closely at them.
Indeed. One of FreeBSD's block device encryption classes (phk's gbde) uses MD5 as a "bit blender" function for instance; just because something's using a hash function doesn't mean it's depending on it to be cryptographically secure; there are plenty of cases where you just want a nice distribution of seed bits to play with, like with BloomFilters and such.
Re:Can someone please explain to me...
on
Opera Free as in Beer
·
· Score: 4, Informative
"built-in bit torrent client"
This is still only in the 8.10/8.02 previews, right? I don't see it in the changelogs or feature lists for 8.50.
Re:Can someone please explain to me...
on
Opera Free as in Beer
·
· Score: 3, Interesting
I use this in Opera, not sure if it works in Firefox (last time I checked this use of content: didn't work there). I've sadly lost the original source, so if anyone recognises it...
body:before { position: absolute; visibility: hidden; background-image: url("javascript:(function(){ window.onclick = function(){ var srcElem=window.event.srcElement; var tag=srcElem.tagName; if(tag=='EMBED') srcElem.className='zichtbaar'}; })()"); content:""}
Add to your user CSS file or edit the.ini to put it in the same list as the bundled CSS files; turns Flash into ickle grey "Flash" boxes you can click on to load.
The only software of theirs we've used other than the drivers bundled with Linux/FreeBSD have been aaccli; a management interface is pretty important with a RAID card:)
The smallest system we've ran them on is a dual 2.8GHz Xeon with 4G of memory; we're currently testing one in a dual 265 (dual dual core 1.8GHz Opteron) with 16G of memory and 8 15kRPM SCSI drives in RAID10, and while it works (albeit throwing Machine Check Exceptions under IO load), performance is merely "sufficient"; Linux md raid10 is generally faster, more flexible, and seems less likely to fail horribly during normal operation, although a proper RAID card with a battery backup module would probably manage power failures better.
It's certainly possible to be lucky with these cards, probably down to picking the right combination of firmware, driver versions, drives, motherboard, IO load, and orientation with the nearest layline. It's a lot more hit-and-miss than I'd like, though; for something that's supposed to increase reliability and uptime, we've probably had more downtime from card failures than drive failures. YMMV.
Intel make MegaRAID cards based on the same design LSI use, btw; may be worth a look if you're not looking at systems with tonnes of memory. They do PCI Express versions too, making them one of the first PCI-E IO cards I've seen.
Many CD-DA copy protection schemes like to munge error correction codes which make CD-ROM drives and ripping software very upset, but which most cheapo CD players don't even notice. These definately break the standards, hence many copy protected CD's lacking the CD-DA logo.
It's just a shame these discs don't have to be clearly labelled by law as not being "real" audio CD's. I basically don't buy CD's any more because it's a crapshoot as to whether you get a real disc or not.
RockBox comes highly recommended, with ports to quite a few newer players being underway, including the iAUDIO ones; it supports Ogg Vorbis, FLAC, and even ReplayGain iirc (features page is down atm) which is great for keeping volume sensible. I'm gonna wait and see how far these ports get before investing in a player.
AAC is "Adaptec AdvancedRAID Controller", and yeah, I just checked and that model is indeed one.
It is indeed a real hardware controller, but it's quite likely they have a bog standard SATA controller behind the 80303; I think they have a AIC79XX behind the SCSI models, but I dunno, I've never had a good look at the physical card.
They're not *awful*; our master database has ran one for years with few problems (FreeBSD; uptime currently 218 days, with ~5.2 billion queries), the main one being disks randomly popping off the array for no apparant reason, which could be drive firmware related. Still, we had enough problems with the controller occasionally stopping processing IO's for no apparant reason (mainly on Linux) to think that maybe they're not the most robust devices on the planet. Google should confirm that these aren't the most problem-free controllers about.
Better card? Well, LSI seem to know what they're doing, but after finding the latest driver doesn't want to talk to our card (maybe because we have >8G of memory) I'm dubious there too, although their hearts seem to be in the right place with plenty of community interaction and support. 3Ware have a lot of people raving about their SATA cards but I have no experience with them.
So, er, yeah, I'd be interested in recommendations too:)
(Currently fighting a 2120S; "No partitionable media were found." says debian-installer. I'll be glad when we're back to software RAID.)
Well, yeah, the sizes of higher end drives tend to be pretty poor because they use smaller, less dense platters to aid seek times (less distance for the head to travel, less time waiting for the head to settle on a track). These drives quote latencies of ~9ms; Raptors and SCSI drives start at around 4.5ms and get as low as 3.
In a server environment where you're limited by the number of drives you can fit in your icke pizzabox (and power and cool effectively) you can't always just say "let's throw twice as many drives at the problem" if it's raw IO's/sec that you're worried about rather than space or serial transfer rate.
They're AAC devices, I believe. If their SCSI models are anything to go by, I'd say.. avoid like the plague. Between buggy firmware, overheating hardware, crappy drivers and even crappier management tools, you're probably better off spending that money on a motherboard with a decent bus (which you'll need to drive any non-trivial RAID card anyway) and just using software RAID on a known-good controller. This is based on several years experience with a number of 2120S cards and various systems and OS's.
Seriously, this is the company which sells a £50 dual port SATA card based on the same SiI chipset you find in £5 cards; that same chipset widely regarded as being one of the buggiest controllers in common use today. I shudder to think what they fit behind their proper SATA RAID controllers.
Quite. 32bit 33MHz PCI (especially shared among on-board stuff *and* multiple card slots) is amazingly feeble these days, so consumer-level PCI Express comes not a minute too soon. Of course if you can afford and appreciate 8 200G drives you can probably also afford and appreciate a half-decent workstation/server board with PCI-X, but even a pair of modern drives can completely saturate the bus, and if you're into file sharing over GigE even one drive is way too much.
For that matter even sharing/dev/zero over GigE on PCI is.. disappointing.
Slashdot Mirror
Blame driver vendors for thinking desktop users don't care. And blame desktop users for generally not caring, or even knowing it's possible for that matter.
Right, these clusters of behaviors and conditions that we've identified and labelled and which you even admit to existing, have no meaning. It's all just part of life's rich tapestry, and that's that.
Similarly, there's no such thing as planets, stars, moons, etc; there's just clusters of matter with various characteristics and states that tend to occur together throughout the observed universe -- the labels we've applied to them are meaningless buzzwords, and in 100 years we'll have moved on to the next stupid faddy names.
And of course since I've never met Bram, and know next to nothing about him beyond one bit of software he wrote and an interview or two he did, I can feel confident in saying that his behavior isn't worthy of a label, even if it might help him and everyone else cope with and understand his problems. They certainly haven't impacted his life in such a profound way that being able to talk about them might hold any importance to him or any of the similarly geeky people likely to be reading his interviews.
What, is there something about certain Rails components that replacing one with something that better suits your requirements (and which is largely API compatible, I might add) will cause the magic smoke to escape?
Yes, Rails has limitations; so does everything else. That's why we design systems that are loosely coupled so components can be swapped out as and when needed. Ruby is especially good at this, and Rails is no exception.
Object Graphs is an alternative to ActiveRecord that has quite flexible support for mapping legacy schemas to your objects. Nitro itself looks like a nice alternative to Rails too, but you can mix and match if you feel like it.
I have no idea, really, I'm just speculating; I suspect it's not as simple as that. They're already hideously complicated and expensive to design; putting two on a chip, or even just two on a card, may well prove to be easier than working out how to integrate 700 million transistors into a single core. *shrug*
:)
I started with the GeForce 256DDR (T&L, woo!), Ti4200, 5900XT, 6800GT and now a 7800GTX. I'm also okay with this
Tyan S2895: ~£320
2*Opteron 265; dual core 1.8GHz: ~£1000
4*1GB ECC Registered memory: ~£500
Nice Supermicro EATX case: ~£300
That's over £2k without storage, hardware RAID, graphics, etc, with a fairly modest amount of wriggle room for further savings. I hope you have a really good use for all those cores, or at least enough money that the cost is largely irrelevent. You save a lot going dual single core, and there are still reasons you might prefer a simple dual Opteron over an X2 (like support for >4G of memory and more readily available PCI-X), but if you're just looking at a nice desktop rather than a professional graphics workstation or big multi-user server, they probably don't apply much in this case.
Personally, I went for a mid-range X2, left the dual dual Opteron in the data centre and am awaiting a handmedown IBM eServer for my home serverish needs. YMMV.
Soft shadows in games like Chronices of Riddick and FEAR really take it out of my 7800GTX, especially at my TFT's native resolution (1600*1200). SLI's probably pretty much the only way to play at such high settings reasonably. Complex maps/situations in other games can also make it chug, and I'm sure it only gets worse at higher resolutions and AA levels (I normally play with 4X AA, an SLI user will probably be breezing along at 16X).
Then of course there's the people who use 3D hardware as part of their job; CAD, 3D artists, level designers, game engine developers; one of the first SLI forum threads I read was by a guy involved in medical imaging. SLI is also laying the groundwork for future multicore cards; in much the same way that SMP has been the realm of rich bastards and high end professional users until multicore consumer level CPU's, SLI will probably remain in the realm of the same sort of people for a year or two until we start seeing multicore NV chips.
Right, that's why I said "come up with novel/fun ways of dealing with the scales involved", not "copy Elite II, it was the best game evar!!11!12".
Elite II (Frontier) had pretty realistic physics; I think it's a shame more games didn't try to run with that, and come up with novel/fun ways of dealing with the huge scales involved, instead of copping out with small environments and mysterious drag keeping everything within a few hundred m/s.
pkill/grep are nice too, and are standard on a fair few systems now.
Well, presumably they've improved somewhat since the cheapy chipsets I played with last year then; they could hardly have gotten worse ;)
:)
;)
At ~30MB/s I see something like 3% CPU use on my X2 4400+ and an on-board PCI-E Marvell; once I get my fileserver sorted out with PCI-X (surplus rackmount server) I'd hope to see more exciting transfer rates more in tune with modern HD's. Presumably similar could be said with two consumer-level PCI-E systems now; that makes me very happy
For comparison, on a 3200+ I see ~30% CPU use on an E1000 and ~80% on a Realtek using similar transfer rates, so you see where my hate for rl comes from
"gigabit PCI cards for $5"
To hell with that; GigE can totally hammer the dinky PCI bus most systems have these days, you want something with proper interrupt moderation, 100% working checksum offload and decent DMA support; jumbo frames are nice too. Cheapy Realtek cards need not apply; spend a bit more on an Intel Pro/1000 or so, unless you like having 80% of your CPU dedicated to servicing interrupts and copying data around pointlessly.
PCI-E stuff seems much better, thankfully; makes on-board GigE on consumer level systems actually quite nice.
Yup, Aliens. The scene in the movie was quite weak so didn't make it to the theatrical release, but you can still see it (among other deleted scenes) in the Special Edition DVD (R2 at least).
I wrote my own ickle ad killer for Opera using CSS, which mostly scratches my itch. I'm sure there are better ones out there, but I haven't looked too closely at them.
Indeed. One of FreeBSD's block device encryption classes (phk's gbde) uses MD5 as a "bit blender" function for instance; just because something's using a hash function doesn't mean it's depending on it to be cryptographically secure; there are plenty of cases where you just want a nice distribution of seed bits to play with, like with BloomFilters and such.
"built-in bit torrent client"
This is still only in the 8.10/8.02 previews, right? I don't see it in the changelogs or feature lists for 8.50.
The only software of theirs we've used other than the drivers bundled with Linux/FreeBSD have been aaccli; a management interface is pretty important with a RAID card :)
The smallest system we've ran them on is a dual 2.8GHz Xeon with 4G of memory; we're currently testing one in a dual 265 (dual dual core 1.8GHz Opteron) with 16G of memory and 8 15kRPM SCSI drives in RAID10, and while it works (albeit throwing Machine Check Exceptions under IO load), performance is merely "sufficient"; Linux md raid10 is generally faster, more flexible, and seems less likely to fail horribly during normal operation, although a proper RAID card with a battery backup module would probably manage power failures better.
It's certainly possible to be lucky with these cards, probably down to picking the right combination of firmware, driver versions, drives, motherboard, IO load, and orientation with the nearest layline. It's a lot more hit-and-miss than I'd like, though; for something that's supposed to increase reliability and uptime, we've probably had more downtime from card failures than drive failures. YMMV.
Intel make MegaRAID cards based on the same design LSI use, btw; may be worth a look if you're not looking at systems with tonnes of memory. They do PCI Express versions too, making them one of the first PCI-E IO cards I've seen.
Many CD-DA copy protection schemes like to munge error correction codes which make CD-ROM drives and ripping software very upset, but which most cheapo CD players don't even notice. These definately break the standards, hence many copy protected CD's lacking the CD-DA logo.
It's just a shame these discs don't have to be clearly labelled by law as not being "real" audio CD's. I basically don't buy CD's any more because it's a crapshoot as to whether you get a real disc or not.
RockBox comes highly recommended, with ports to quite a few newer players being underway, including the iAUDIO ones; it supports Ogg Vorbis, FLAC, and even ReplayGain iirc (features page is down atm) which is great for keeping volume sensible. I'm gonna wait and see how far these ports get before investing in a player.
Well, thanks for that hugely convincing, well thought out and detailed rational argument. Which two-word reply do you think you've earned?
Please try again, preferably with less handwaving. I'd be especially interested in MySQL benchmarks, if you have them, btw.
AAC is "Adaptec AdvancedRAID Controller", and yeah, I just checked and that model is indeed one.
:)
It is indeed a real hardware controller, but it's quite likely they have a bog standard SATA controller behind the 80303; I think they have a AIC79XX behind the SCSI models, but I dunno, I've never had a good look at the physical card.
They're not *awful*; our master database has ran one for years with few problems (FreeBSD; uptime currently 218 days, with ~5.2 billion queries), the main one being disks randomly popping off the array for no apparant reason, which could be drive firmware related. Still, we had enough problems with the controller occasionally stopping processing IO's for no apparant reason (mainly on Linux) to think that maybe they're not the most robust devices on the planet. Google should confirm that these aren't the most problem-free controllers about.
Better card? Well, LSI seem to know what they're doing, but after finding the latest driver doesn't want to talk to our card (maybe because we have >8G of memory) I'm dubious there too, although their hearts seem to be in the right place with plenty of community interaction and support. 3Ware have a lot of people raving about their SATA cards but I have no experience with them.
So, er, yeah, I'd be interested in recommendations too
(Currently fighting a 2120S; "No partitionable media were found." says debian-installer. I'll be glad when we're back to software RAID.)
Well, yeah, the sizes of higher end drives tend to be pretty poor because they use smaller, less dense platters to aid seek times (less distance for the head to travel, less time waiting for the head to settle on a track). These drives quote latencies of ~9ms; Raptors and SCSI drives start at around 4.5ms and get as low as 3.
In a server environment where you're limited by the number of drives you can fit in your icke pizzabox (and power and cool effectively) you can't always just say "let's throw twice as many drives at the problem" if it's raw IO's/sec that you're worried about rather than space or serial transfer rate.
They're AAC devices, I believe. If their SCSI models are anything to go by, I'd say.. avoid like the plague. Between buggy firmware, overheating hardware, crappy drivers and even crappier management tools, you're probably better off spending that money on a motherboard with a decent bus (which you'll need to drive any non-trivial RAID card anyway) and just using software RAID on a known-good controller. This is based on several years experience with a number of 2120S cards and various systems and OS's.
Seriously, this is the company which sells a £50 dual port SATA card based on the same SiI chipset you find in £5 cards; that same chipset widely regarded as being one of the buggiest controllers in common use today. I shudder to think what they fit behind their proper SATA RAID controllers.
Quite. 32bit 33MHz PCI (especially shared among on-board stuff *and* multiple card slots) is amazingly feeble these days, so consumer-level PCI Express comes not a minute too soon. Of course if you can afford and appreciate 8 200G drives you can probably also afford and appreciate a half-decent workstation/server board with PCI-X, but even a pair of modern drives can completely saturate the bus, and if you're into file sharing over GigE even one drive is way too much.
/dev/zero over GigE on PCI is.. disappointing.
For that matter even sharing