You're kidding, right? It guarantees that the few remaining insurgent groups will prepare for the date, and then attack with whatever they have left. That's why you *don't have a specific date* nor do you release your plans to the enemy.
Of course. And I am certainly not claiming the absurd notion that the G4 is a better processor, or generically faster. I haven't spent a lot of effort to figure out why other than Altivec. On MATLAB benchmarks, which is a little more legitimate test, with nothing else running aside from the OS in both cases, the calculations benchmarks are faster on the Dual G4 than the Core 2 Duo machine, but the graphics are much, much slower to the point the overall benchmark rating is at the dead bottom of the list. I actually don't care that much, I can do massively parallel processing by running the PC, the G4, and the Sun workstation via Exceed on the PC, all at the same time. I don't get 3x the pay, however.
My point was, however, that the thing is hardly the paperweight slug that everybody was suggesting. And that's a G4 - that we bought about a month ago, I might add! Still need Classic for other things. And, until about a month ago we were using an 8600/300, which was no speed demon but adequate to the task. The only reason we aren't still using it is that I had to run 10.xxx to be allowed to connect it to the PC network, and the hard drive was too small to contain the OS. It was cheaper to get a $200 G4 MDD than it was a copy of OS10.4 and a hard drive.
I went and checked the various G5 models and those calling for it to be recycled are loonies. We are talking ~$500 worth of machine there, maybe a lot more for the higher-zoot version.
I think you encapsulate the problem nicely. I didn't say that it wasn't important or useful for certain applications, but we are indeed working directly on the bit level- or at least should. Lots and Lots of people might die if this thing craps out under stressed conditions. Until recently, no one even understood how the exception handling worked in the OS, (or, really, how OS worked at all beyond the broadest brush for that matter), and *I* was the one to discover that the way the code was written, *exceptions were not detectable*. For the last 10+ years, and nobody noticed. Of course it was never tested because it's hard to create the exception in the first place. That places people's life in danger.
Any bug I made when I was writing it putting your life in mortal danger *right now* - which is why we insisted on writing clearly understandable code, checking it down the bit level (yes, high level code testing, examine assembly code, dump as hex, compare that bits matched assembly code using a processor handbook) and then single-stepped through it with a debugger checking that the registers did what we expected. Found things like a "MOV" instruction that worked in most of the conditions, but not when some register bit that shouldn't have mattered was set. There's now 4 pages of errata with one-liners listing them for our processor - something like 200 of them. So much for "trusting the tools". We could test it do that degree because we could easily see what it was doing and knew exactly what each bit of data should have looked like - and it was easy. We made HUGE modifications with great ease because it was laid out at a high level in so straightforward way that anybody could just jump in and do it. And still do.
I understand that this can't be done in large commercial projects and that the cost/benefit ratio is a real factor and how good you want to design/test it VS how much cost and schedule is a perfectly legitimate tradeoff. And this sort of programming may be the ideal solution to this problem. But when it's treated like the One True Way, and the result is that it's completely unmaintainable, mostly untestable, and the authors of the code consider it so fragile that they insist that it never, ever, be touched, in what should be a pretty small life-critical system, that's when I have a big problem with it.
Not really. I can write in Java and C++ just fine, thank you. If your concept of a "design paradigm" is to write code that even the authors don't understand and are terrified of even slightly changing for fear of making it not work any more, then I will pass. My code had been easily maintainable for about 2 decades now/
Hey, I am using a Wind Tunnel G4 right now. It works fantastically well and is lightning fast at running satellite simulations. Faster than my Core 2 Duo PC by a fair margin.
These sorts of languages (and the underlying religious cults they bring with them) are probably appropriate for some uses. But what I see done in my life-critical real-time processor applications borders on criminal. Data hiding? How the f'ing hell do I check what is going on to the bit level is some twit determined to "hide the data". This is particularly apt right now, because we are adding a feature to our code that was almost trivial to add to our FORTRAN simulations, and because of the "cult of classes" C++ programming it's damn near impossible in the final product, and completely impossible to look at and tell what the heck it's doing. Trying to test it like a black box is never going to get to the level we need.
We started having peer reviews of the code, and my colleagues and I are the designers of the system, so we would hypothetically need to sign off on it. We went for two hours to get 10 lines into it, no one could explain how it was working but that we should just "trust the compiler". That didn't fly with us, so the solution was to *not have us present at the peer reviews* since we were "disruptive"
What we need is someone that can write straightforward procedural code, but no one seem to be willing or able to do it any more. It has all the features of a cult or a secret society, even when you get someone to understand and agree, they won't deviate from their dogma.
No, actually, it's precisely the opposite. The other films (Velvia, primarily) are favored precisely because they distort reality. Velvia is particularly inaccurate - take a picture of one of your Hispanic friends and see what the skin tone looks like. Then do the same with Kodachrome, and with a Nikon DSLR. The Nikon will be almost perfectly accurate, the Kodachrome will be almost as good with the very slightest greenish cast, and the Velvia will look like you spray-painted her face with Candy-Apple Red car paint.
Kodachrome was arguably the most accurate slide film. The problem is, digital color accuracy is better than any film, and people who still shoot slide film are doing it for the artistic qualities, not documentation, and like and can take advantage of the distortion of reality provided by Velvia. Reality is generally quite boring, artistically.
Another factor, not relevant to the average amateur, is that Velvia is available in a variety of formats 35mm, 120, 4x5 (and maybe 8x10) so you can get the same results in many formats. Also, the processing of Velvia is far cheaper and E6 processing has been far easier and more available for 30 years. That's the only thing that kept Ektachrome alive over the years - for most of its history, it has been useless crap, and was never the first choice, used only when you had to get it processed quickly or wanted to do it at home.
Not even close to correct. The equipment is essentially unique and not at all like any other processing system. That was always one of the issues - there was never going to be anything like "1-hour processing" for Kodachrome, the process is two orders of magnitude more difficult and fussy than anything else.
It was incredibly stable, the colors were very well-saturated but otherwise pretty accurate. The last version (kodachrome 64) was a little too contrasty for my tastes - I liked Kodachrome 25. You can't compare it to any digital until your pixels get smaller than a silver molecule, at least not any "35 mm" digital camera. Used to us it in 120 and just looking at them on a light table made you feel like a hero. "Kodachrome Red" was pretty famous, red always looked great. And it was perfectly well-suited for skin tones.
The film that effectively put it out of business, Fuji Velvia, is spectacular for landscapes where it pumps up the color saturation, and has all the colors like Kodachrome did red. It's very warm as far as color balance goes, and skin tones are almost cartoonishly shifted. It's essentially useless for portraits because of that. But it's far easier to process and you can still get it processed locally. Dwayne's Photo has been the only place processing it for years, and if you took it to a shop, that;s were it eventually ended up. Typically in recent years the turnaround time is on the order of two weeks. E6, you can still get overnight.
As far as I have seen, there's no real general-purpose replacement for Kodachrome. OR, rather, its digital - where the lack of image quality is offset by far superior color accuracy (much better on a general basis than ANY film) and easily manipulated and printed images.
But the handwriting is on the wall for just about all 35mm. It's always been marginally acceptable for sports and photojournalism because it was cheap and the little cameras were reasonably portable. The lack of overall image quality compared to 120 or larger (other than in the hands of masters) didn't really matter for magazines or newspapers. But everything 35mm could do is more-or-less easier or better with digital aside from the image quality, and the image quality of digital (since the mid 00's) has been sufficient to the point that it didn't matter.
When I go on photo trips, I now carry 4 cameras - a Canon point-and-shoot for quickies, a Nikon digital SLR for anything that moves, and two Yashica-Mats, one with Velvia 100 and one with Tri-X, for things that don't move. I will typically take the same shot with the Velvia 100 and the Nikon just in case, and meter the Yashica-mat shots with the Nikon (to back up spot meter readings).
BTW, if you get out the OM-1, be sure and check the foam light seals on the back. I have a 1977 version and the foam is decaying severely. and bear in mind that you can't get the batteries for the meter any more - they make some replacements but most of them don't put out the right voltage.
The Senate unanimously agreed to do something that is actually good?
Of course, they are basically saying that US priorities take precedence over foreign priorities, so I suppose it doesn't really count.
You think that is an unusual position for someone that is charged with representing the people of the State of Alabama? Of course they should consider the priorities of the constituents should take precedence, that's their job and the job of the US Government.
Of course, "troll" is in the eye of the beholder. Slashdot's system enforces a monoculture of thought as restrictive as any I have seen on the internet. Now maybe that what people want and it's moderately democratic in the way it is done, but to claim it's a bastion of free speech and acceptance of varying opinions and perspective is a huge misrepresentation.
What the heck are you talking about? A square foot, say, is a measure of area. Scale the linear dimensions and the value of the area goes up by the square of the linear dimensions. A square foot = 144 square inches. Same thing with european units, presumably.
Slashdot Mirror
No kidding! Far fewer people died of H1N1 than died of conventional yearly flu - by several orders of magnitude.
I was very surprised at how many people *do* die of conventional flu every year.
Talk about fanbois - it gores your ox so it can't possibly be true? Any basis aside from that for thinking it's not correct?
Three is the number of the counting, and the counting of the number shall be three.
You're kidding, right? It guarantees that the few remaining insurgent groups will prepare for the date, and then attack with whatever they have left. That's why you *don't have a specific date* nor do you release your plans to the enemy.
If you are going to put that kind of thought into cheating on a test, wouldn't you be better served actually learning the material?
Oh bullshit - it's never going to be built, and the money will be pissed away
What's your point? We are also committed to building a high-speed train from Barstow to Lodi, at astonishing cost.
Of course. And I am certainly not claiming the absurd notion that the G4 is a better processor, or generically faster. I haven't spent a lot of effort to figure out why other than Altivec. On MATLAB benchmarks, which is a little more legitimate test, with nothing else running aside from the OS in both cases, the calculations benchmarks are faster on the Dual G4 than the Core 2 Duo machine, but the graphics are much, much slower to the point the overall benchmark rating is at the dead bottom of the list. I actually don't care that much, I can do massively parallel processing by running the PC, the G4, and the Sun workstation via Exceed on the PC, all at the same time. I don't get 3x the pay, however.
My point was, however, that the thing is hardly the paperweight slug that everybody was suggesting. And that's a G4 - that we bought about a month ago, I might add! Still need Classic for other things. And, until about a month ago we were using an 8600/300, which was no speed demon but adequate to the task. The only reason we aren't still using it is that I had to run 10.xxx to be allowed to connect it to the PC network, and the hard drive was too small to contain the OS. It was cheaper to get a $200 G4 MDD than it was a copy of OS10.4 and a hard drive.
I went and checked the various G5 models and those calling for it to be recycled are loonies. We are talking ~$500 worth of machine there, maybe a lot more for the higher-zoot version.
I think you encapsulate the problem nicely. I didn't say that it wasn't important or useful for certain applications, but we are indeed working directly on the bit level- or at least should. Lots and Lots of people might die if this thing craps out under stressed conditions. Until recently, no one even understood how the exception handling worked in the OS, (or, really, how OS worked at all beyond the broadest brush for that matter), and *I* was the one to discover that the way the code was written, *exceptions were not detectable*. For the last 10+ years, and nobody noticed. Of course it was never tested because it's hard to create the exception in the first place. That places people's life in danger.
Any bug I made when I was writing it putting your life in mortal danger *right now* - which is why we insisted on writing clearly understandable code, checking it down the bit level (yes, high level code testing, examine assembly code, dump as hex, compare that bits matched assembly code using a processor handbook) and then single-stepped through it with a debugger checking that the registers did what we expected. Found things like a "MOV" instruction that worked in most of the conditions, but not when some register bit that shouldn't have mattered was set. There's now 4 pages of errata with one-liners listing them for our processor - something like 200 of them. So much for "trusting the tools". We could test it do that degree because we could easily see what it was doing and knew exactly what each bit of data should have looked like - and it was easy. We made HUGE modifications with great ease because it was laid out at a high level in so straightforward way that anybody could just jump in and do it. And still do.
I understand that this can't be done in large commercial projects and that the cost/benefit ratio is a real factor and how good you want to design/test it VS how much cost and schedule is a perfectly legitimate tradeoff. And this sort of programming may be the ideal solution to this problem. But when it's treated like the One True Way, and the result is that it's completely unmaintainable, mostly untestable, and the authors of the code consider it so fragile that they insist that it never, ever, be touched, in what should be a pretty small life-critical system, that's when I have a big problem with it.
Not really. I can write in Java and C++ just fine, thank you. If your concept of a "design paradigm" is to write code that even the authors don't understand and are terrified of even slightly changing for fear of making it not work any more, then I will pass. My code had been easily maintainable for about 2 decades now/
No, because the Core 2 Duo is a Windows PC.
On the core 2 duo, it takes about 10 minutes for a 3-day run, on the G4 it takes about 3.5 minutes.
Brett
Hey, I am using a Wind Tunnel G4 right now. It works fantastically well and is lightning fast at running satellite simulations. Faster than my Core 2 Duo PC by a fair margin.
These sorts of languages (and the underlying religious cults they bring with them) are probably appropriate for some uses. But what I see done in my life-critical real-time processor applications borders on criminal. Data hiding? How the f'ing hell do I check what is going on to the bit level is some twit determined to "hide the data". This is particularly apt right now, because we are adding a feature to our code that was almost trivial to add to our FORTRAN simulations, and because of the "cult of classes" C++ programming it's damn near impossible in the final product, and completely impossible to look at and tell what the heck it's doing. Trying to test it like a black box is never going to get to the level we need.
We started having peer reviews of the code, and my colleagues and I are the designers of the system, so we would hypothetically need to sign off on it. We went for two hours to get 10 lines into it, no one could explain how it was working but that we should just "trust the compiler". That didn't fly with us, so the solution was to *not have us present at the peer reviews* since we were "disruptive"
What we need is someone that can write straightforward procedural code, but no one seem to be willing or able to do it any more. It has all the features of a cult or a secret society, even when you get someone to understand and agree, they won't deviate from their dogma.
This will help you sort it out - they are in it for the money! That's not a bad thing, thats what they are there for.
No, actually, it's precisely the opposite. The other films (Velvia, primarily) are favored precisely because they distort reality. Velvia is particularly inaccurate - take a picture of one of your Hispanic friends and see what the skin tone looks like. Then do the same with Kodachrome, and with a Nikon DSLR. The Nikon will be almost perfectly accurate, the Kodachrome will be almost as good with the very slightest greenish cast, and the Velvia will look like you spray-painted her face with Candy-Apple Red car paint.
Kodachrome was arguably the most accurate slide film. The problem is, digital color accuracy is better than any film, and people who still shoot slide film are doing it for the artistic qualities, not documentation, and like and can take advantage of the distortion of reality provided by Velvia. Reality is generally quite boring, artistically.
Another factor, not relevant to the average amateur, is that Velvia is available in a variety of formats 35mm, 120, 4x5 (and maybe 8x10) so you can get the same results in many formats. Also, the processing of Velvia is far cheaper and E6 processing has been far easier and more available for 30 years. That's the only thing that kept Ektachrome alive over the years - for most of its history, it has been useless crap, and was never the first choice, used only when you had to get it processed quickly or wanted to do it at home.
Not even close to correct. The equipment is essentially unique and not at all like any other processing system. That was always one of the issues - there was never going to be anything like "1-hour processing" for Kodachrome, the process is two orders of magnitude more difficult and fussy than anything else.
It was incredibly stable, the colors were very well-saturated but otherwise pretty accurate. The last version (kodachrome 64) was a little too contrasty for my tastes - I liked Kodachrome 25. You can't compare it to any digital until your pixels get smaller than a silver molecule, at least not any "35 mm" digital camera. Used to us it in 120 and just looking at them on a light table made you feel like a hero. "Kodachrome Red" was pretty famous, red always looked great. And it was perfectly well-suited for skin tones.
The film that effectively put it out of business, Fuji Velvia, is spectacular for landscapes where it pumps up the color saturation, and has all the colors like Kodachrome did red. It's very warm as far as color balance goes, and skin tones are almost cartoonishly shifted. It's essentially useless for portraits because of that. But it's far easier to process and you can still get it processed locally. Dwayne's Photo has been the only place processing it for years, and if you took it to a shop, that;s were it eventually ended up. Typically in recent years the turnaround time is on the order of two weeks. E6, you can still get overnight.
As far as I have seen, there's no real general-purpose replacement for Kodachrome. OR, rather, its digital - where the lack of image quality is offset by far superior color accuracy (much better on a general basis than ANY film) and easily manipulated and printed images.
But the handwriting is on the wall for just about all 35mm. It's always been marginally acceptable for sports and photojournalism because it was cheap and the little cameras were reasonably portable. The lack of overall image quality compared to 120 or larger (other than in the hands of masters) didn't really matter for magazines or newspapers. But everything 35mm could do is more-or-less easier or better with digital aside from the image quality, and the image quality of digital (since the mid 00's) has been sufficient to the point that it didn't matter.
When I go on photo trips, I now carry 4 cameras - a Canon point-and-shoot for quickies, a Nikon digital SLR for anything that moves, and two Yashica-Mats, one with Velvia 100 and one with Tri-X, for things that don't move. I will typically take the same shot with the Velvia 100 and the Nikon just in case, and meter the Yashica-mat shots with the Nikon (to back up spot meter readings).
BTW, if you get out the OM-1, be sure and check the foam light seals on the back. I have a 1977 version and the foam is decaying severely. and bear in mind that you can't get the batteries for the meter any more - they make some replacements but most of them don't put out the right voltage.
You think that is an unusual position for someone that is charged with representing the people of the State of Alabama? Of course they should consider the priorities of the constituents should take precedence, that's their job and the job of the US Government.
Brett
Shouldn't that be +1 Tasteless?
So, to double check, count them!
Of course, "troll" is in the eye of the beholder. Slashdot's system enforces a monoculture of thought as restrictive as any I have seen on the internet. Now maybe that what people want and it's moderately democratic in the way it is done, but to claim it's a bastion of free speech and acceptance of varying opinions and perspective is a huge misrepresentation.
I always suspected that Windows amounted to sabotage!
What the heck are you talking about? A square foot, say, is a measure of area. Scale the linear dimensions and the value of the area goes up by the square of the linear dimensions. A square foot = 144 square inches. Same thing with european units, presumably.
It's the Judean People's Front crack Suicide Team to the rescue!