Practical example: the latency to the NTP server I sync to randomly varies between 30ms and 70ms. The range of my offset varies from roughly -5ms to +5ms. Pretty accurate, all things considered.
Most applications don't need hyper-accurate precision. There are some use-cases out there where it might matter, but for the vast majority of workloads, a few milliseconds of time offset isn't a problem, even in enterprise.
I have a couple of thousand physical servers. They very much need to sync their hardware clocks via. NTP. I need reliable NTP servers. NTP running on a virtual host is not reliable (the clock drifts horribly, although ESX5i is better in this regard).
A host running ntpd against an external source won't drive at all, virtual or otherwise. Presumably you have to run your own NTP server because you want to save the transit costs of all your servers making requests to external NTP servers. In that case, one of your servers would be synchronized to the root pool, for example, and the rest would synchronize to that one.
If you're running an NTP server that is itself not synchronized to a higher level external NTP server or some physical external clock, that NTP server is going to drift even if it's running on real hardware.
So, yeah, they had one crappy film, and two "merely good" films, but they've done a lot of fantastic work since they were bought out by Disney. Cars was mostly made before Disney took over, and after that, there was an amazing run...
I can even envision one of the teasers for the unmolested high def release.
Simply show the most infamous scene, remastered in high def but otherwise unmodified from the theatrical release... with only one shot fired, as it should be. You end the teaser as the scene normally ends: Han flips the money at the bartender, says "Sorry about the mess", and the screen cuts to a simple date on a pure black background.
They end episode 6 with the scrappy rebel alliance lopping the head off a vast galactic empire. Such a thing doesn't instantly disappear just because the person running it is gone.
Counterpoint: Empire Strikes Back, widely regarded as the best of the six films, was made by Irvin Kershner in his late 50s towards the end of his career. At the time, he'd been directing movies for over twenty years.
And yet the sales marketshare isn't reflected in tablet browser marketshare, which says some interesting things about how many people continue using their tablets after purchase.
Somebody who owns a smartphone, laptop, and tablet technically has their smart devices outpacing their PCs two to one, but they probably still use their PC for the same stuff they did before...
I love my smartphone, but it's replacing my feature phone, not my desktop.
When I bought my mac a few months ago, I had a list of software that I wanted to install. I had done my research, and so I knew what apps I wanted to start with. You know, I wanted Chrome as a browser, Cord as a remote desktop client, etc. Not one single piece of software that I wanted was available in the app store.
From a user's perspective, getting stuff from the app store is nice since it unifies the update mechanisms, but it's not useful if there's no content.
Estimates put over 700,000 on Apple's appstore. Even if only 20% are active, that's still 140k, which is a great deal more than the 7000 (of which a similar percentage rule likely applies) available on Surface.
I don't really get Microsoft here. They have this opportunity to use the strength of Windows openness as a development platform (in that anybody can develop for it without approval or fees) to do to tablets what Android did to smartphones (dominate the market). Instead, they're locking the thing down even tighter than Apple, and even pissing developers off by reserving a few rather blatant privileges for themselves (only Microsoft is allowed to developer RT desktop apps)...
Microsoft is the newcomer to an established market here (Tablet PC was a non-starter), they're not going to make much headway trying to out-restrict Apple. Android only supplanted the iPhone because they took the opposite approach.
Reviews indicate that single-threaded performance is a huge bottleneck for the Surface (Microsoft Office maxes out a CPU core just when you type text), so the fact that the Tegra 3 (with its Cortex A9s) is being used everywhere (and at a low clockspeed to boot) is a big problem. Had they gone with even a dual core Krait, (which at reasonable clockspeeds can have more than double the single-threaded performance) they would have been much better off.
Indeed, and I was tempted to post a comment about how researchers had rediscovered forward error correction. This technique is already used extensively absolutely everywhere, from optical media to DSL lines. What's so new and impressive about this "new" version of FEC?
They did, since PowerPC was originally more of a requirement spec than a processor, and Apple was the one setting those requirements... The process that took the PowerPC from being what it was before to what it was after was defined by the customer.
Customer (Apple) says I need a processor with these characteristics, IBM adapts one of their existing products to fit the customer's needs. Not sure what's so ominous about that.
You've got a rather lot of exaggeration and hyperbole there. For example, Apple never claimed to have singlehandedly created the PowerPC, but they did create it in concert with IBM and Motorola as part of the AIM alliance, and they did initiate the creation process. There was no PowerPC before Apple set the ball rolling; PowerPC was an adaptation of IBM's existing POWER architecture, specifically a single-chip version of the POWER1 processor. Apple never stabbed Exponential Technology in the back by deciding at the last minute not to use their CPUs, they were never going to, since ET's chips underperformed and Motorola convinced Apple not to switch. Etc.
Or perhaps Samsung was simply not willing to reduce prices as low as Apple's other screen manufacturers like LG were? Or was not willing to commit to the volumes Apple wanted? Or any other many reasons why they might end this kind of supply contract?
That's a last-gen part, a Sandy Bridge. It would be slower in everything but heavily multithreaded applications, since it has slightly lower IPC and clockspeed.
I'm confused, the D and E bridges were completely different in appearance, I don't know why you're bringing that in. The Vegas recreation bridge and the TV filming bridge are rather different. The doors on the Vegas bridge are much wider (for practical reasons), but many other shapes of things are off (curves are wrong, lengths are wrong). For one example, the seats on the filming set had a wide gap down the middle a few inches wide, while on the Vegas chairs (which are a rather different shape) there isn't a hole there at all, just a seam...
This has nothing to do with BluRay (which didn't make any changes to the set, no clue what you're talking about) or DVD or the E set. The Vegas set was done from scratch and wasn't very accurate, plain and simple.
Slashdot Mirror
Practical example: the latency to the NTP server I sync to randomly varies between 30ms and 70ms. The range of my offset varies from roughly -5ms to +5ms. Pretty accurate, all things considered.
Most applications don't need hyper-accurate precision. There are some use-cases out there where it might matter, but for the vast majority of workloads, a few milliseconds of time offset isn't a problem, even in enterprise.
I have a couple of thousand physical servers. They very much need to sync their hardware clocks via. NTP. I need reliable NTP servers. NTP running on a virtual host is not reliable (the clock drifts horribly, although ESX5i is better in this regard).
A host running ntpd against an external source won't drive at all, virtual or otherwise. Presumably you have to run your own NTP server because you want to save the transit costs of all your servers making requests to external NTP servers. In that case, one of your servers would be synchronized to the root pool, for example, and the rest would synchronize to that one.
If you're running an NTP server that is itself not synchronized to a higher level external NTP server or some physical external clock, that NTP server is going to drift even if it's running on real hardware.
Pirates of Penzance is fantastic, but it has no connection to Disney... Perhaps you meant "Pirates of the Caribbean"?
The Black Hole was made in 1979... Most of the decision makers of the time are long gone.
As for Pixar, well, look at the full list of Pixar filmes released after Disney bought them (under the "Disney-Pixar" brand)
Cars (2006, 74% RT)
Ratatouille (2007, 96% RT)
WALL-E (2008, 96% RT)
Up (2009, 98% RT)
Toy Story 3 (2010, 99% RT)
Cars 2 (2011, 38% RT)
Brave (2012, 78% RT)
So, yeah, they had one crappy film, and two "merely good" films, but they've done a lot of fantastic work since they were bought out by Disney. Cars was mostly made before Disney took over, and after that, there was an amazing run...
Why, you think Joss Whedon's evil twin brother Josh Whedon would do a better job?
He only really made one of the original three movies. Empire? Jedi? He neither wrote nor directed those ones.
I can even envision one of the teasers for the unmolested high def release.
Simply show the most infamous scene, remastered in high def but otherwise unmodified from the theatrical release... with only one shot fired, as it should be. You end the teaser as the scene normally ends: Han flips the money at the bartender, says "Sorry about the mess", and the screen cuts to a simple date on a pure black background.
Better cardboard than CG. You're talking about the sets, right? :P
They end episode 6 with the scrappy rebel alliance lopping the head off a vast galactic empire. Such a thing doesn't instantly disappear just because the person running it is gone.
No, The Dude is only 62.
And the recent Muppet movie turned out pretty good too...
Counterpoint: Empire Strikes Back, widely regarded as the best of the six films, was made by Irvin Kershner in his late 50s towards the end of his career. At the time, he'd been directing movies for over twenty years.
And yet the sales marketshare isn't reflected in tablet browser marketshare, which says some interesting things about how many people continue using their tablets after purchase.
Somebody who owns a smartphone, laptop, and tablet technically has their smart devices outpacing their PCs two to one, but they probably still use their PC for the same stuff they did before...
I love my smartphone, but it's replacing my feature phone, not my desktop.
When I bought my mac a few months ago, I had a list of software that I wanted to install. I had done my research, and so I knew what apps I wanted to start with. You know, I wanted Chrome as a browser, Cord as a remote desktop client, etc. Not one single piece of software that I wanted was available in the app store.
From a user's perspective, getting stuff from the app store is nice since it unifies the update mechanisms, but it's not useful if there's no content.
Estimates put over 700,000 on Apple's appstore. Even if only 20% are active, that's still 140k, which is a great deal more than the 7000 (of which a similar percentage rule likely applies) available on Surface.
I don't really get Microsoft here. They have this opportunity to use the strength of Windows openness as a development platform (in that anybody can develop for it without approval or fees) to do to tablets what Android did to smartphones (dominate the market). Instead, they're locking the thing down even tighter than Apple, and even pissing developers off by reserving a few rather blatant privileges for themselves (only Microsoft is allowed to developer RT desktop apps)...
Microsoft is the newcomer to an established market here (Tablet PC was a non-starter), they're not going to make much headway trying to out-restrict Apple. Android only supplanted the iPhone because they took the opposite approach.
Reviews indicate that single-threaded performance is a huge bottleneck for the Surface (Microsoft Office maxes out a CPU core just when you type text), so the fact that the Tegra 3 (with its Cortex A9s) is being used everywhere (and at a low clockspeed to boot) is a big problem. Had they gone with even a dual core Krait, (which at reasonable clockspeeds can have more than double the single-threaded performance) they would have been much better off.
Except Apple joined AIM four years before CHRP...
Indeed, and I was tempted to post a comment about how researchers had rediscovered forward error correction. This technique is already used extensively absolutely everywhere, from optical media to DSL lines. What's so new and impressive about this "new" version of FEC?
They did, since PowerPC was originally more of a requirement spec than a processor, and Apple was the one setting those requirements... The process that took the PowerPC from being what it was before to what it was after was defined by the customer.
Customer (Apple) says I need a processor with these characteristics, IBM adapts one of their existing products to fit the customer's needs. Not sure what's so ominous about that.
You've got a rather lot of exaggeration and hyperbole there. For example, Apple never claimed to have singlehandedly created the PowerPC, but they did create it in concert with IBM and Motorola as part of the AIM alliance, and they did initiate the creation process. There was no PowerPC before Apple set the ball rolling; PowerPC was an adaptation of IBM's existing POWER architecture, specifically a single-chip version of the POWER1 processor. Apple never stabbed Exponential Technology in the back by deciding at the last minute not to use their CPUs, they were never going to, since ET's chips underperformed and Motorola convinced Apple not to switch. Etc.
Or perhaps Samsung was simply not willing to reduce prices as low as Apple's other screen manufacturers like LG were? Or was not willing to commit to the volumes Apple wanted? Or any other many reasons why they might end this kind of supply contract?
That's a last-gen part, a Sandy Bridge. It would be slower in everything but heavily multithreaded applications, since it has slightly lower IPC and clockspeed.
I'm confused, the D and E bridges were completely different in appearance, I don't know why you're bringing that in. The Vegas recreation bridge and the TV filming bridge are rather different. The doors on the Vegas bridge are much wider (for practical reasons), but many other shapes of things are off (curves are wrong, lengths are wrong). For one example, the seats on the filming set had a wide gap down the middle a few inches wide, while on the Vegas chairs (which are a rather different shape) there isn't a hole there at all, just a seam...
This has nothing to do with BluRay (which didn't make any changes to the set, no clue what you're talking about) or DVD or the E set. The Vegas set was done from scratch and wasn't very accurate, plain and simple.