The way AMD has designed these smaller "packages" to work together as one CPU is, to put it very simply, to have them communicate through the DDR4 bus.
There won't be any Alienware setups that are CPU bound in gaming. These will be attached to either high resolution or high frame rate monitors (or both) and will be GPU bound. And you'll also get the benefit of multiple cores for everything else.
Started out with Redhat 6.2, long before there was a Fedora. Never had to reinstall to upgrade, even when they renamed it Fedora.
Back then we called it "Fedora Core":
Before Fedora 7, Fedora was called Fedora Core after the name of one of the two main software repositories - Core and Extras. Fedora Core contained all the base packages that were required by the operating system, as well as other packages that were distributed along with the installation CD/DVDs, and was maintained only by Red Hat developers. Fedora Extras, the secondary repository that had been included since Fedora Core 3, was community-maintained and not distributed along with the installation CD/DVDs. Upon the release of Fedora 7, the distinction between Fedora Core and Fedora Extras was eliminated.[35]
I started out with Redhat around 4.0 (mid 90s) and never looked back. I've tried out many distributions over the years but for me it's always provided a really great compromise in features and stability.
I am, but I wouldn't say it's solely based on inertia. I regularly assess what distro I'm running on my workstation.
I think the release cadence (6 months) provides a good balance between features and stability. I also use RHEL/CentOS professionally so being able to transfer skills between the two is very helpful (file system layout, RPM, etc). I don't see any distribution doing anything significantly better than Fedora. Possibly Ubuntu but I don't think there would be any significant gains for a lot of re-learning.
Never required a $2500 PC (more like $500-$1000, you know, like any decent gaming PC). Tons of available games. I love watching all the old slashdot luddites shit on VR, most of whom have never experienced it.
I think it's illustrating all the people who had been waiting for the Ryzen launch. There was significant pent up demand for AMD parts, but not for Intel. So I would have expected a huge surge in AMD sales. The question will be how well they do long term. I'm personally just waiting for the Threadripper announcement before I decide to upgrade to Ryzen or Threadripper.
That's what I mean by "braindead". A big deadline should not translate to OT, and OT is never a good fix for bad planning.
You're telling me working an extra 5 hours one week a year to meet a deadline because client requirements changed, or Suzie was out sick, or some kind of operational problem poppedup, or one of a million other unforeseen circumstances is "bad planning" ? I guess we'll just have to agree to disagree.
How do you manage Global Address List and Distribution Lists using something other than Exchange? Serious question because that's the one I always get stuck on when I dream of ditching Exchange.
Braindead manager here. I think that's really a misreading of the situation. What we value is people who put in the time when there's really a need (ie big deadline, critical failure, etc). If someone is working 50, 60 or more hours per week every week, I'm starting to get worried. You really don't want someone who's that committed to burn themselves out.
Conclusion
On paper, the Zenbook Pro is an incredible deal. It features some extremely high-end parts, yet is amazingly affordable compared to its competitors.
After spending time with it, however, it becomes clear why it's less expensive, particularly when compared to the Dell XPS 15 with the exact same specs. The Zenbook Pro'ss performance is slightly weaker, and it has subpar speakers and little oddities like uneven keyboard backlighting and that slap-dash Thunderbolt sticker.
If all you care about is GPU, RAM, CPU and HDD then do not buy a Mac. If you're willing to put up with awful keyboards and trackpads and substandard displays to save some money, there are better options. Honestly, I'm not being sarcastic.
systemd can either be modular or "one place you need to secure". So, systemd proponents, which is it?
How it's structured between modules when there are no alternatives to those modules and they are all interdependent on one another makes that a philosophical debate, not a practical one.
What systemd does is create that socket so that Service X can bind to it right away, and proceed with startup, and once Y arrives at the point of creation of the socket, it's bound to the socket Systemd created. There, instead of sequence of complete startups of services, which is lengthy, you have only sequence of socket binding, which is fast, and parallel startup, which is fast too.
I actually just read this chapter in the RHCSA book I was reading (don't give me that look, I just wanted to see if there were any gaps to fill in). It was the one idea I wasn't aware of that actually made a lot of sense. One of the first things it does is initialize all those sockets, and if an application tries to communicate before the service actually connects to the socket, it just holds it in a buffer. That's actually a pretty cool idea!
Unfortunately systemd is an overgrown mess that's slowly snaking it's way through everything. If this is what the open source community decides is best, fine, I'm just along for the ride, but I genuinely feel like this is an overwrought very unnecessary system that eventually linux, in the general sense, will become far too dependent on.
but smaller customers that moved to the "cloud" and have experienced too many failures, legal issues or increased scrutiny are scaling back what is effectively "shared hosting" back to self-hosting or at least colocating their own 'cloud'.
I don't think the numbers back up the claim. If you look at the massive growth I think you find exactly the opposite. I'm sure there are anecdotal stories but the stats just don't bear that out.
Has anyone honestly ever been happy with the headaches of moving to Office365 or AWS (beyond the higher ups)?
No point giving you anecdotal examples, I think we just have to look at the numbers.
Do you have a source to back up your assertion that these customers are leaving these platforms in any significant numbers?
I think it may just be a matter of terminology. It's kind of unfair to combine every branch of the military and every contractor compared to a single company. Or not, but I guess it just matters on how you want to define a single "consumer". The article mentions "company" before using the "consumer" description. Of course timing is also important, I still can't find any numbers from the DoD that show more than that currently in place or even a projected timeline.
URL doesn't work, but fair enough. When will more than 400,000 be on Windows 10? The only source I can find says they won't even start until late 2016. So this may just be a question of timing.
Interestingly enough, the US DoD is only 450,000 and I'm willing to bet the vast majority aren't even close to being on Windows 10. I think if you added in all the third party contractors it would be much larger, but those would be separate companies.
I thought there were lots of reports of infections in Russia? Seems like a dangerous move.
But in general I think Russia's flagrant hacking is really going to come back to bite them. I believe the US is much better at this than Russia. And even if you disagree with that, I don't think any reasonable person would disagree that the US plus its major allies (ie Canada, UK, Germany, etc) are vastly better at this.
I think the only argument you could make is, well they're already attacking Russia and now Russia is just very publicly fighting back. Maybe in an attempt to position themselves to have negotiations for a "truce" between all nations. I think Russia would benefit a lot more from an agreement than the US would, so maybe appearing to be the largest threat actor helps their negotiating position?
Those exceptions are so rare people are still writing about how unusual it is that Dropbox left the public cloud with headlines like "THE EPIC STORY OF DROPBOX'S EXODUS FROM THE AMAZON CLOUD EMPIRE". This is only worth doing at extremely large scale. And, as the competition between providers (AWS, Azure, Google Cloud) continues and their scale increase, prices continue to be driven down which means fewer and fewer customers are large enough to build these platforms themselves less expensively.
I think we'll continue to see lots of hybrid clouds because right now the big providers aren't worried about building for unusual edge-case workloads. Right now their best return is focusing on moving generalized workloads as quickly into public clouds to drive growth and ultimately their scale to reduce their cost. Once we see that growth slow down we'll start to see more specialized offerings. Which is why you'll still see hybrid clouds being very popular in the short to mid-term.
Slashdot Mirror
The way AMD has designed these smaller "packages" to work together as one CPU is, to put it very simply, to have them communicate through the DDR4 bus.
I thought they communicated via Infinity Fabric?
There won't be any Alienware setups that are CPU bound in gaming. These will be attached to either high resolution or high frame rate monitors (or both) and will be GPU bound. And you'll also get the benefit of multiple cores for everything else.
node.js and ruby puppy mills
Started out with Redhat 6.2, long before there was a Fedora. Never had to reinstall to upgrade, even when they renamed it Fedora.
Back then we called it "Fedora Core":
Before Fedora 7, Fedora was called Fedora Core after the name of one of the two main software repositories - Core and Extras. Fedora Core contained all the base packages that were required by the operating system, as well as other packages that were distributed along with the installation CD/DVDs, and was maintained only by Red Hat developers. Fedora Extras, the secondary repository that had been included since Fedora Core 3, was community-maintained and not distributed along with the installation CD/DVDs. Upon the release of Fedora 7, the distinction between Fedora Core and Fedora Extras was eliminated.[35]
I started out with Redhat around 4.0 (mid 90s) and never looked back. I've tried out many distributions over the years but for me it's always provided a really great compromise in features and stability.
I am, but I wouldn't say it's solely based on inertia. I regularly assess what distro I'm running on my workstation.
I think the release cadence (6 months) provides a good balance between features and stability. I also use RHEL/CentOS professionally so being able to transfer skills between the two is very helpful (file system layout, RPM, etc). I don't see any distribution doing anything significantly better than Fedora. Possibly Ubuntu but I don't think there would be any significant gains for a lot of re-learning.
Never required a $2500 PC (more like $500-$1000, you know, like any decent gaming PC). Tons of available games. I love watching all the old slashdot luddites shit on VR, most of whom have never experienced it.
Your scores are questionable. Passmark shows a ~20% better overall performance from the 1400 for a CPU that costs $4 more.
I think it's illustrating all the people who had been waiting for the Ryzen launch. There was significant pent up demand for AMD parts, but not for Intel. So I would have expected a huge surge in AMD sales. The question will be how well they do long term. I'm personally just waiting for the Threadripper announcement before I decide to upgrade to Ryzen or Threadripper.
Sounds like they just "pivoted" from the Umbrella rental business to the umbrella selling business. They got deposits.
Just use this: https://www.bankofamerica.com/...
That's what I mean by "braindead". A big deadline should not translate to OT, and OT is never a good fix for bad planning.
You're telling me working an extra 5 hours one week a year to meet a deadline because client requirements changed, or Suzie was out sick, or some kind of operational problem poppedup, or one of a million other unforeseen circumstances is "bad planning" ? I guess we'll just have to agree to disagree.
How do you manage Global Address List and Distribution Lists using something other than Exchange? Serious question because that's the one I always get stuck on when I dream of ditching Exchange.
Braindead manager here. I think that's really a misreading of the situation. What we value is people who put in the time when there's really a need (ie big deadline, critical failure, etc). If someone is working 50, 60 or more hours per week every week, I'm starting to get worried. You really don't want someone who's that committed to burn themselves out.
http://www.pcworld.com/article...
Conclusion On paper, the Zenbook Pro is an incredible deal. It features some extremely high-end parts, yet is amazingly affordable compared to its competitors. After spending time with it, however, it becomes clear why it's less expensive, particularly when compared to the Dell XPS 15 with the exact same specs. The Zenbook Pro'ss performance is slightly weaker, and it has subpar speakers and little oddities like uneven keyboard backlighting and that slap-dash Thunderbolt sticker.
If all you care about is GPU, RAM, CPU and HDD then do not buy a Mac. If you're willing to put up with awful keyboards and trackpads and substandard displays to save some money, there are better options. Honestly, I'm not being sarcastic.
systemd can either be modular or "one place you need to secure". So, systemd proponents, which is it?
How it's structured between modules when there are no alternatives to those modules and they are all interdependent on one another makes that a philosophical debate, not a practical one.
What systemd does is create that socket so that Service X can bind to it right away, and proceed with startup, and once Y arrives at the point of creation of the socket, it's bound to the socket Systemd created. There, instead of sequence of complete startups of services, which is lengthy, you have only sequence of socket binding, which is fast, and parallel startup, which is fast too.
I actually just read this chapter in the RHCSA book I was reading (don't give me that look, I just wanted to see if there were any gaps to fill in). It was the one idea I wasn't aware of that actually made a lot of sense. One of the first things it does is initialize all those sockets, and if an application tries to communicate before the service actually connects to the socket, it just holds it in a buffer. That's actually a pretty cool idea!
Unfortunately systemd is an overgrown mess that's slowly snaking it's way through everything. If this is what the open source community decides is best, fine, I'm just along for the ride, but I genuinely feel like this is an overwrought very unnecessary system that eventually linux, in the general sense, will become far too dependent on.
Remind me again, how many remote remote code execution bugs were in sysvinit? I can't remember.
but smaller customers that moved to the "cloud" and have experienced too many failures, legal issues or increased scrutiny are scaling back what is effectively "shared hosting" back to self-hosting or at least colocating their own 'cloud'.
I don't think the numbers back up the claim. If you look at the massive growth I think you find exactly the opposite. I'm sure there are anecdotal stories but the stats just don't bear that out.
Has anyone honestly ever been happy with the headaches of moving to Office365 or AWS (beyond the higher ups)?
No point giving you anecdotal examples, I think we just have to look at the numbers.
Do you have a source to back up your assertion that these customers are leaving these platforms in any significant numbers?
I think it may just be a matter of terminology. It's kind of unfair to combine every branch of the military and every contractor compared to a single company. Or not, but I guess it just matters on how you want to define a single "consumer". The article mentions "company" before using the "consumer" description. Of course timing is also important, I still can't find any numbers from the DoD that show more than that currently in place or even a projected timeline.
URL doesn't work, but fair enough. When will more than 400,000 be on Windows 10? The only source I can find says they won't even start until late 2016. So this may just be a question of timing.
Interestingly enough, the US DoD is only 450,000 and I'm willing to bet the vast majority aren't even close to being on Windows 10. I think if you added in all the third party contractors it would be much larger, but those would be separate companies.
I thought there were lots of reports of infections in Russia? Seems like a dangerous move.
But in general I think Russia's flagrant hacking is really going to come back to bite them. I believe the US is much better at this than Russia. And even if you disagree with that, I don't think any reasonable person would disagree that the US plus its major allies (ie Canada, UK, Germany, etc) are vastly better at this.
I think the only argument you could make is, well they're already attacking Russia and now Russia is just very publicly fighting back. Maybe in an attempt to position themselves to have negotiations for a "truce" between all nations. I think Russia would benefit a lot more from an agreement than the US would, so maybe appearing to be the largest threat actor helps their negotiating position?
What about when people make two directly contradictory general statements? I was making a guess as to what you meant, apparently I was right.
Those exceptions are so rare people are still writing about how unusual it is that Dropbox left the public cloud with headlines like "THE EPIC STORY OF DROPBOX'S EXODUS FROM THE AMAZON CLOUD EMPIRE". This is only worth doing at extremely large scale. And, as the competition between providers (AWS, Azure, Google Cloud) continues and their scale increase, prices continue to be driven down which means fewer and fewer customers are large enough to build these platforms themselves less expensively.
I think we'll continue to see lots of hybrid clouds because right now the big providers aren't worried about building for unusual edge-case workloads. Right now their best return is focusing on moving generalized workloads as quickly into public clouds to drive growth and ultimately their scale to reduce their cost. Once we see that growth slow down we'll start to see more specialized offerings. Which is why you'll still see hybrid clouds being very popular in the short to mid-term.
Sure, Amazon has some pet food sales. But people aren't ever going to stop buying dog food locally.
These two sentences seem to directly contradict one another. Do you mean that SOME people will always buy it locally?