I see a lot of these "Software programmers don't understand this" posts, on newsgroups etc. I found verilog easy to understand, and I was very much a software programmer. Perhaps in days of yore when C programmers had simple single-threaded code it was appropriate, but the application my team is currently working on frequently has 100 threads or so concurrently running. Constructing an 'always@' block around some signal *is* pretty similar to how we multithread programs these days...
Once you've started coding parallel apps, it's hard to go back to serial ones - I liken it to when you wrote procedural code and then moved to object-orientated code. It's possible to revert, but not pleasant. When that's your mindset, picking up either of the two languages isn't that hard (VHDL is far too verbose for me, personally, though).
The concepts are simple (if fundamental and sometimes subtle), the syntax is simple, conceptualising it is also pretty simple; given that most of my designs have been relatively tiny compared to the software projects I work on (the current project is some 900k LOC). My most-complex verilog project was a real-time HD JPEG-2k encoder. Lots of parallelism and pipelining to get the frame-rate:)
Oh, and I'm hearing good things about SystemC - which seems to take some of the lower-level boring work out of HDL coding. Probably worth looking at too...
Well, there's the rub though; company employees they don't get to assume something and then physically threaten people in a public place, even if those people are doing something the employees don't like.
And police(wo)men don't get to arrest you just because they want to. If there was any law that had been broken, the photographer would have been charged with it by the police once he'd been taken to the station. If nothing else, it makes it harder for him to sue their ass. As he ought to be doing right now.
They can whine and bleat on about what they assumed, or what they suspected, or what they thought was happening, and frankly it's all irrelevant. The *documented* *actions* of both the police and the employees were illegal, pure and simple.
If what he says is true, he ought to be suing them for emotional distress, unlawful arrest, unlawful detainment, and threatening behaviour. He also ought to be seeking punitive relief against the police for their part in it. Gaining money isn't the reason here, it's the way to call attention to the behaviour of the officer in question.
Or, for that matter, would be breaking by taking a photo of the interior of an arnoured car ?
I'm not suggesting that in either case the owners/operators would be happy about it, but the job of the police is to enforce the law, not to make one group happy at the expense of another (unless the two groups are law-abiding-citizens and criminals, of course:)....
Actually at peak, they register ~7kW between them all (as read off the converter directly). Figure the desert boosting that to ~10 and we end up somewhere in-between at 86,400 * 6 or half a million panels.
That's 4460/mile, or roughly 1 every 14 inches as the guy below said. Figure 5 lengthwise across the track and it's easily possible, but you'd have to use those new-fangled light-concentrating panels to make it even vaguely financially viable, rather than my old-fashioned ones:)
I have 48 solar panels on my roof in northern CA. Yesterday they generated 45 kWH between them. Figure that the middle of the desert is actually a better solar energy source and bump that to (say) 60, and the multiplier becomes 110,000 / 60 = ~1800 times as many panels or 86,400.
There's ~116 miles between Tucson and Phoenix. That's ~750 panels per mile. It's a lot, but it's not unfeasible.
I'm not saying your concerns aren't valid, I think some of them are, but the energy side could be made to work. They ought to get a significant discount on the price (~ $1k/panel) if they're ordering circa 90,000 of them, which should help their cost-benefit analysis:)
If indeed, it is "impossible to predict earthquakes, it seems to me that getting a minor quake on-the-day of prediction, and the major quake hitting a week later is pretty much as good as could possibly have been expected.
Now if all he did was guess, it'd be a whole different ball-game, but as far as I remember, doing this "science" thingy involves recognising a problem, taking measurements, postulating a theory to fit those measurements, and (sadly, in this case) testing that theory against further predictions it made. Seems like he followed the rule-book on that one...
Part of the problem, of course, is that people (including, one might say *especially*, elected officals) aren't good at assessing risk. They consider risk to be the consequences of an event, whereas really it's the consequences of an event multiplied by the probability of that event. It's why we look out for "global killer" meteorites, even though they are incredibly unlikely. The risk inherent in such a strike makes it worthwhile to keep putting in the effort at detecting them. It's easiest to illustrate when the fate of the whole world lies in balance, but the principle remains the same even for localised disasters such as this one...
So often, it comes down to better education being the key to good decision-making. Why is it that we let people who only want to run for power take on the mantle of power over us ? I recall a Sci-Fi story where on election, all a (wo)man's worldly goods were forcibly sold, and the cash amount held in trust. Once the successor appeared, the departing official was given access to his/her trust fund again - the implication being that you had to do well by everyone else before you could do well for yourself. I'm not suggesting this is workable, but perhaps an element of personal stake might be a useful thing for a politician to have... Perhaps then they'd listen to the scientist, and not just go on gut instinct...
Looks like I wasn't obvious enough - see my earlier comment on the swallowing of < in the post. Earlier than both of these cowardly responses, in fact....
I think you just proved the point that you didn't read my parent's post... I was originally responding to:
Who in the world is this LaCie external drive made for?
As for your comparison
The G-Raid mini isn't Apple-specific. The device has usb-2 as well as firewire 4/800 ports.
The one you suggest is 9.5" x 6" x 3" (compared to 5.875" x 3.25" x 1.5") or ~6x as bulky
Yours is not bus-powered, so I can't take it into the field. Not many AC power-points halfway up a volcano when you need to clear space on the CF cards...
It comes from the manufacturer of drives which have served me well in the past, not the cheapest knock-down price you can find on the internet. It looks and feels well-engineered, and they have performed very well in very demanding environments for around a year now.
In short, your choice is useless to me. This isn't a matter of thinking 'Apple products are "worth it"', it's a matter of finding something that's good for the job. I don't see any problem with that.
As for "paranoia", if I'm sent on a shoot, the cost of an extra drive is negligible - literally lost in the noise - and it may not even be possible to re-obtain the photograph. If it's the Beijing olympics that you're covering, you can't really ask them to do it again if your hard-drive dies... People seem to object to having to get married again as well...
Pretty much any serious software allows you to copy from one card to multiple destinations - and by your definition using that feature makes most of the photographers I know "paranoid". Hmm, perhaps they're all wrong and you're right. Perhaps.
Even the lowly Mac mini is a dual-core system. Every laptop is a dual-core system. The Mac Pro is either 4-core (with hyperthreading for a virtual 8-core) or 8-core (with hyperthreading for a virtual 16-core) system.
"Better to keep silent and look the fool, rather than speak and remove all doubt"
How does using a netbook help me store huge quantities of large RAW images when using Aperture ? Are you seriously suggesting I use a networked filesystem, or something ? If you are, I'm speechless.
Even if it weren't a stupid idea, using *one* drive (and an internal one at that!) is an amateur's mistake. You copy the photos off the card simultaneously to at least two hard drives (that's two mini-G's, not one mini-G with two disks inside). You burn the images to optical (DVD, Blu-ray,...) soon afterwards.
The mini-G is firewire-800, it's plenty fast enough, and it mates nicely with the Macbook Pro that's running Aperture.
Because each photo is ~70MB, and I take several thousand on a shoot. Leaf digital backs are fantastic, but the photos take up space. Thank [insert deity] for the way Aperture handles image-adjustments...
I want speed (which two RAID drives gives me) because transferring all those files from cards to dual disks, and then actually rough-cutting them is a major time sink. The Mini-G's have firewire-800 as their interface, so you get to see the benefits of both disks.
I want capacity (which two RAID drives gives me) for all the obvious reasons.
I don't need long-term reliability - all the photos are mirrored onto two of these mini-G drives, and then burnt onto DVD by my assistant when I get back anyway.
And if you use it like I use my G-Raid-mini, taking it with you as the storage for your portable when on photography trips, you'd be the one with the backache, and I'd be strolling around with.5T in my jacket pocket.
Same old fallacy: Just because *I* can't use it means that *you* can't use it. Just not true...
*EVERY* time Apple announce something new for the touch/iphone, it costs an extra $10 on the touch.
*EVERY* time someone moans about that.
*EVERY* time someone else points out that Apple account for iphone sales over a period of time, thus allowing them to maneuver around the ridiculous Sarbonnes-Oxley requirements. They bill the touch as a one-off, so can't add new functionality without there being a representative charge.
Whether you agree with them or not, that's their position (presumably that of their highly-paid lawyers, too). Get over it, *every* time you add onto the touch, you're going to pay extra.
From the *summary* (for [insert deity]'s sake man, at least read the *summary*) of the 'most ungreen companies ever' link you gave above:
"Ars Technica points out that Greenpeace's research isn't quite up-to-snuff, and it's also worth noting that Greenpeace admitted to targeting Apple for the publicity in the past."
... they wouldn't be able to claim it, unless they had some justification for it. From what I read, Greenpeace don't really care about what you *do* these days, they care about what you *promise* to do in the future, and how much you pay them to be quiet. They're a form of eco-terrorists, and eventually they'll get theirs...
Get a 'Striker' laser-target-enabled missile launcher ($40, I think). Then get a webcam or IP-enabled camera (I got one of these from Ebay for ~$70).
Use the camera to detect motion and generate a centroid of motion; use the (high-intensity of red) laser-spot to detect where the missile is pointed (again from the camera image), and move the missile to make the centroid and laser-spot coincident.
It's actually pretty trivial, but it looks pretty cool to have people walk into the office and have two missile-launchers automatically track them.
I also have the think-geek big-red-button box, which I modified to allow the button to control a USB port. Now I can fire the (auto-targetting:) missiles by hitting the big-red-button:)
It's actually only slightly harder to get the system to track two independent targets... The next step is to build in target-recognition by accessing the company's person-directory (we all have pictures)... Don't shoot the VP. Only directors and below are valid targets:)
I believe that's the purpose behind 'Snow Leopard'... Apple have taken a time-out to work on the underpinnings of the OS, and tune-up the performance.
The two things I'm most looking forward to are OpenCL (a standard way to access any GPU from user-code, AFAICT) and Grand Central (a way to easily harness multiple CPU's in a standardised way. Between the two of those, I can see Apple leaping ahead in performance over "normal" PC's...
From my perspective, I see it as Apple sweating the details, so I don't have to. They have a history of doing that, and I for one appreciate it.
Slashdot Mirror
Snap.
Agreed.
I see a lot of these "Software programmers don't understand this" posts, on newsgroups etc. I found verilog easy to understand, and I was very much a software programmer. Perhaps in days of yore when C programmers had simple single-threaded code it was appropriate, but the application my team is currently working on frequently has 100 threads or so concurrently running. Constructing an 'always@' block around some signal *is* pretty similar to how we multithread programs these days...
Once you've started coding parallel apps, it's hard to go back to serial ones - I liken it to when you wrote procedural code and then moved to object-orientated code. It's possible to revert, but not pleasant. When that's your mindset, picking up either of the two languages isn't that hard (VHDL is far too verbose for me, personally, though).
The concepts are simple (if fundamental and sometimes subtle), the syntax is simple, conceptualising it is also pretty simple; given that most of my designs have been relatively tiny compared to the software projects I work on (the current project is some 900k LOC). My most-complex verilog project was a real-time HD JPEG-2k encoder. Lots of parallelism and pipelining to get the frame-rate :)
Oh, and I'm hearing good things about SystemC - which seems to take some of the lower-level boring work out of HDL coding. Probably worth looking at too...
Simon
Well, there's the rub though; company employees they don't get to assume something and then physically threaten people in a public place, even if those people are doing something the employees don't like.
And police(wo)men don't get to arrest you just because they want to. If there was any law that had been broken, the photographer would have been charged with it by the police once he'd been taken to the station. If nothing else, it makes it harder for him to sue their ass. As he ought to be doing right now.
They can whine and bleat on about what they assumed, or what they suspected, or what they thought was happening, and frankly it's all irrelevant. The *documented* *actions* of both the police and the employees were illegal, pure and simple.
If what he says is true, he ought to be suing them for emotional distress, unlawful arrest, unlawful detainment, and threatening behaviour. He also ought to be seeking punitive relief against the police for their part in it. Gaining money isn't the reason here, it's the way to call attention to the behaviour of the officer in question.
Simon.
Exactly which law do you think he was breaking ?
Or, for that matter, would be breaking by taking a photo of the interior of an arnoured car ?
I'm not suggesting that in either case the owners/operators would be happy about it, but the job of the police is to enforce the law, not to make one group happy at the expense of another (unless the two groups are law-abiding-citizens and criminals, of course :) ....
Simon.
Actually at peak, they register ~7kW between them all (as read off the converter directly). Figure the desert boosting that to ~10 and we end up somewhere in-between at 86,400 * 6 or half a million panels.
That's 4460/mile, or roughly 1 every 14 inches as the guy below said. Figure 5 lengthwise across the track and it's easily possible, but you'd have to use those new-fangled light-concentrating panels to make it even vaguely financially viable, rather than my old-fashioned ones :)
Simon.
Hmm. Perhaps.
I have 48 solar panels on my roof in northern CA. Yesterday they generated 45 kWH between them. Figure that the middle of the desert is actually a better solar energy source and bump that to (say) 60, and the multiplier becomes 110,000 / 60 = ~1800 times as many panels or 86,400.
There's ~116 miles between Tucson and Phoenix. That's ~750 panels per mile. It's a lot, but it's not unfeasible.
I'm not saying your concerns aren't valid, I think some of them are, but the energy side could be made to work. They ought to get a significant discount on the price (~ $1k/panel) if they're ordering circa 90,000 of them, which should help their cost-benefit analysis :)
Simon.
Well actually not I, but Apple have - see this patent filing story
In fact, they mention using air as the actuator, as far back as 2007.
If indeed, it is "impossible to predict earthquakes, it seems to me that getting a minor quake on-the-day of prediction, and the major quake hitting a week later is pretty much as good as could possibly have been expected.
Now if all he did was guess, it'd be a whole different ball-game, but as far as I remember, doing this "science" thingy involves recognising a problem, taking measurements, postulating a theory to fit those measurements, and (sadly, in this case) testing that theory against further predictions it made. Seems like he followed the rule-book on that one...
Part of the problem, of course, is that people (including, one might say *especially*, elected officals) aren't good at assessing risk. They consider risk to be the consequences of an event, whereas really it's the consequences of an event multiplied by the probability of that event. It's why we look out for "global killer" meteorites, even though they are incredibly unlikely. The risk inherent in such a strike makes it worthwhile to keep putting in the effort at detecting them. It's easiest to illustrate when the fate of the whole world lies in balance, but the principle remains the same even for localised disasters such as this one...
So often, it comes down to better education being the key to good decision-making. Why is it that we let people who only want to run for power take on the mantle of power over us ? I recall a Sci-Fi story where on election, all a (wo)man's worldly goods were forcibly sold, and the cash amount held in trust. Once the successor appeared, the departing official was given access to his/her trust fund again - the implication being that you had to do well by everyone else before you could do well for yourself. I'm not suggesting this is workable, but perhaps an element of personal stake might be a useful thing for a politician to have... Perhaps then they'd listen to the scientist, and not just go on gut instinct...
Simon.
X11 is a separate download/install on Mac so it's not a realman's Unix.
I haven't downloaded or installed X separately for a long long time. It comes built-in as far as I'm concerned, along with all the development tools.
Simon
Fairy nuff :)
Looks like I wasn't obvious enough - see my earlier comment on the swallowing of < in the post. Earlier than both of these cowardly responses, in fact....
Simon
Gaah - the < was swallowed in the statement "Apple have no <2 core intel systems. Period."
Probably obvious, but to save people nit-picking
I think you just proved the point that you didn't read my parent's post... I was originally responding to:
As for your comparison
In short, your choice is useless to me. This isn't a matter of thinking 'Apple products are "worth it"', it's a matter of finding something that's good for the job. I don't see any problem with that.
As for "paranoia", if I'm sent on a shoot, the cost of an extra drive is negligible - literally lost in the noise - and it may not even be possible to re-obtain the photograph. If it's the Beijing olympics that you're covering, you can't really ask them to do it again if your hard-drive dies... People seem to object to having to get married again as well...
Pretty much any serious software allows you to copy from one card to multiple destinations - and by your definition using that feature makes most of the photographers I know "paranoid". Hmm, perhaps they're all wrong and you're right. Perhaps.
Simon.
Apple have no 2 core intel systems. Period.
Even the lowly Mac mini is a dual-core system. Every laptop is a dual-core system. The Mac Pro is either 4-core (with hyperthreading for a virtual 8-core) or 8-core (with hyperthreading for a virtual 16-core) system.
"Better to keep silent and look the fool, rather than speak and remove all doubt"
Simon.
Are you fecking serious ? That's not even wrong!
How does using a netbook help me store huge quantities of large RAW images when using Aperture ? Are you seriously suggesting I use a networked filesystem, or something ? If you are, I'm speechless.
Even if it weren't a stupid idea, using *one* drive (and an internal one at that!) is an amateur's mistake. You copy the photos off the card simultaneously to at least two hard drives (that's two mini-G's, not one mini-G with two disks inside). You burn the images to optical (DVD, Blu-ray,...) soon afterwards.
The mini-G is firewire-800, it's plenty fast enough, and it mates nicely with the Macbook Pro that's running Aperture.
Simon.
Because each photo is ~70MB, and I take several thousand on a shoot. Leaf digital backs are fantastic, but the photos take up space. Thank [insert deity] for the way Aperture handles image-adjustments...
I want speed (which two RAID drives gives me) because transferring all those files from cards to dual disks, and then actually rough-cutting them is a major time sink. The Mini-G's have firewire-800 as their interface, so you get to see the benefits of both disks.
I want capacity (which two RAID drives gives me) for all the obvious reasons.
I don't need long-term reliability - all the photos are mirrored onto two of these mini-G drives, and then burnt onto DVD by my assistant when I get back anyway.
Simon.
And if you use it like I use my G-Raid-mini, taking it with you as the storage for your portable when on photography trips, you'd be the one with the backache, and I'd be strolling around with .5T in my jacket pocket.
Same old fallacy: Just because *I* can't use it means that *you* can't use it. Just not true...
Simon
I've been using a G-RAID mini for a year or so. The drive I have is only 500GB, but it's fast (for a portable drive) because of the RAID.
There's a 1TB drive coming out soon - see the 'mini-2', which looks to be $699 before any discount (I got ~25% on the mini IIRC).
G-Raid is also a *lot* more reliable than Lacie, in my experience but I guess YMMV, one view is not statistically relevant etc. etc.
Simon
Filtering is best done server side. For me the to-do list is:
Flash
Hell, no.
Java
Not in a million years, thanks.
Printing
Um, perhaps. I never carry a printer around with me, but I can see some use for this
Record video from the camera
That'd be cool.
*EVERY* time Apple announce something new for the touch/iphone, it costs an extra $10 on the touch.
*EVERY* time someone moans about that.
*EVERY* time someone else points out that Apple account for iphone sales over a period of time, thus allowing them to maneuver around the ridiculous Sarbonnes-Oxley requirements. They bill the touch as a one-off, so can't add new functionality without there being a representative charge.
Whether you agree with them or not, that's their position (presumably that of their highly-paid lawyers, too). Get over it, *every* time you add onto the touch, you're going to pay extra.
Simon.
From the *summary* (for [insert deity]'s sake man, at least read the *summary*) of the 'most ungreen companies ever' link you gave above:
... they wouldn't be able to claim it, unless they had some justification for it. From what I read, Greenpeace don't really care about what you *do* these days, they care about what you *promise* to do in the future, and how much you pay them to be quiet. They're a form of eco-terrorists, and eventually they'll get theirs...
As for Darwin, it seems pretty open to me.
Um, like this one ?
Simon
I've lost count of the number of times I've been asked that.... Hmmm....
For the record: not related to Simon T. Perhaps there's something in the name-thing though [grin] ...
Simon.
Get a 'Striker' laser-target-enabled missile launcher ($40, I think). Then get a webcam or IP-enabled camera (I got one of these from Ebay for ~$70).
Use the camera to detect motion and generate a centroid of motion; use the (high-intensity of red) laser-spot to detect where the missile is pointed (again from the camera image), and move the missile to make the centroid and laser-spot coincident.
It's actually pretty trivial, but it looks pretty cool to have people walk into the office and have two missile-launchers automatically track them.
I also have the think-geek big-red-button box, which I modified to allow the button to control a USB port. Now I can fire the (auto-targetting :) missiles by hitting the big-red-button :)
It's actually only slightly harder to get the system to track two independent targets... The next step is to build in target-recognition by accessing the company's person-directory (we all have pictures)... Don't shoot the VP. Only directors and below are valid targets :)
Simon
You're right about that - Apple do indeed have a history of working within the law to try and advance their corporate agenda.
Microsoft, on the other hand, have a history of breaking the law to try and advance their corporate agenda.
I'm not sure how adhering to the law is "more evil" than plain out breaking it, but I guess there's more than beauty in the eye of the beholder.
Simon.
I believe that's the purpose behind 'Snow Leopard'... Apple have taken a time-out to work on the underpinnings of the OS, and tune-up the performance.
The two things I'm most looking forward to are OpenCL (a standard way to access any GPU from user-code, AFAICT) and Grand Central (a way to easily harness multiple CPU's in a standardised way. Between the two of those, I can see Apple leaping ahead in performance over "normal" PC's...
From my perspective, I see it as Apple sweating the details, so I don't have to. They have a history of doing that, and I for one appreciate it.
Simon.