Yah. There's good reasons why a minutes-long delay is not great, because we want things to settle out relatively quickly... but 2.5-25 milliseconds of fuzz on the order book would be just fine.
Why do you still insist it's physically impossible, given that lots of people used it and liked it 7 years ago on connections that were half the speed and double the latency of what's commonly available now? "Physically impossible" is an extraordinary claim. 500 miles at the speed of light is 2.5 milliseconds.
All I'm saying is I played it and I liked it, but games were going to be too uneconomical to justify it. That's pretty much what most of the reviews said, too. This seems to rule out "physically impossible" on the face, because a physical instance actually existed, when consumer internet connections had like one third the speed and double the last mile latency that they have today. Likewise, PS Now's current market success seems to rule out "physically impossible".
Also, you're a dipshit--- you can hardly shill for something that no longer exists.:P
I agree it's fair, but it's not smart. If an employer is still interested in you and you've moved on, no need to alienate them by just ghosting-- if you say "thanks, but this isn't a fit" or "I've found something else"-- it means that whatever positive impression you've created with them can possibly still be useful to you in the future instead of creating a few people who feel the opposite.
That's just not how it went. And in any case, physical limits are hardly preventing PS Now from finding success and getting good reviews-- though people still say you're way better off with ethernet than wifi presently. Even that isn't a fundamental physical limit.
Onlive scoring good (not perfect) reviews for game performance in 2010-2011, as demonstrated below, with much worse than current consumer internet connections... is evidence that we're not up against physical limits. But it was an awful deal, and that + the Microsoft litigation killed them. After the fire sale of assets the service was indeed awful, from what I understand.
Top 3 reviews of found googling the words onlive 2011:
https://www.engadget.com/2010/... > With an up-to-18Mbps AT&T U-Verse connection in San Jose, California, we found OnLive games loaded as quickly as on console -- sometimes much quicker -- and were actually quite playable. The controller never felt quite as responsive as that of a dedicated console nor the images quite as crisp, but we'd say that most of the time the overall experience was only slightly behind what we expect, only bogged down by the occasional annoying stutter. Frantic first-person shooters and driving games weren't as accurate as we like, but over the course of a couple days we adjusted to the mild lag, racking up plenty of kills, scoring the occasional headshot and drifting around some fairly tight corners as well. In Prince of Persia, a game that can require fairly precise timing in combat, we were still able to parry foes' swords and execute tricky jumps with a little bit of forethought, and a multiplayer game of Unreal Tournament III was intriguingly balanced -- if slightly laggy -- thanks to the fact that all players had 0 ping to the (virtual) host server.
https://www.pcgamer.com/onlive... > And yet streaming from the net via OnLive is remarkably playable. Obviously it feels a bit sluggish compared with playing on your own native hardware, but for many games, especially those designed with laggy console controllers in mind, including the likes of Arkham Asylum and Human Revolution, it's far from unpleasant.
http://www.businessinsider.com... > The game had minimal loading times, and while the graphics weren't as crystal clear as on a video game console (because of OnLive's compression technology), the level of detail was pretty amazing. It looks just as good as watching Netflix streaming. Controls originally felt a little delayed, but after a few minutes I felt right at home. I wanted to notice latency and laggy controls (due to my input getting beamed to the over the web, then a response getting beamed back), but I didn't find any in this game.
I'm not shilling for it. I bet at the time it would fail. It was an *awful* deal for games.
Even now, PS Now is doing alright, despite a much smaller deployed footprint (and worse user experience/average latency).
The biggest problems aren't the speed-of-light latency increase (physics) but things like other people in your house using the internet connection (bufferbloat/no reasonable QOS in most consumer edge routers).
Plenty of people play with TVs that add 100ms of latency. 20ms of round trip latency plus 10ms of encoder latency is not the end of the world.
I played Onlive for a month, it was reasonably good. The problems were A) massive capital burn from needing to run datacenters everywhere, and B) shitty deals with publishers, so the economics of buying on Online were awful.
It doesn't take *that* many datacenters to be within 20ms of 90% of the US population.
... Synthesis tools already exist, and every fab has a design library of standard transistors. While the tools are complicated and very expensive (though open source versions exist), they are there. So the problem you're describing is already solved. Designers describe logic, help a little with floorplan and constraints, and get a design out minutes to hours later.
What's difficult is that we don't have great programming mechanisms to describe parallel logic, or to synthesize sequential descriptions of tasks into efficient sets of gates and transistors. Verilog/VHDL/even SystemC are very, very, very slow and cumbersome to develop in and require a shit-ton of verification. If these could be made even a little more like developing software, it would be a big win.
It's reasonable to expect that naughty code can be contained-- we have process containment and virtual machines and OS privileges for the purpose. But side channels, like Spectre etc, make that more difficult.
I've been talking about this a lot lately-- the refereeing in the NBA attracts a lot of complaints.
Right now there's just way too much stuff for referees to look for, at once. There are some things that computers could do very well (3 seconds in the key, lane violations, calling player/ball in/out of bounds, backcourt violation, midcourt 8 second timer). This would free up referees to look for traveling, off-ball contact, etc, so that the overall quality of officiating could improve.
Unless there's some kind of systemic bias --- in what kind of customers someone gets (e.g. ratings are poorer at night)--- it should average out and you should be able to use an average of different customers' types of ratings to compare employees. I agree yes or no questions, and long form "please explain" have value *also* in teasing apart what the customer experience is actually like.
You're missing the fact that 90-95% of humans have HHV6/HHV7, which is what the article is about, which are herpes viruses but not the cause of oral or genital herpes.
Hey, I'm a big Apple fan. The thing that's unfortunate right now is: if you are at a point where you should upgrade systems--- laptop life, OS support, etc--- all of the offerings are underwhelming: dated and not price-performant. Apple has always been a premium option but you'd usually get premium, up to date hardware for it until the past few years.
I'm on a Linux laptop these days and I hope they fix it so I can go back to everyday use of MacOS.
Your population density overall is lower because there's a lot of unpopulated land, but 75% of your population lives in areas with more than 1000 people per square kilometer:P.
The US has telecom problems beyond the population density one, but the bad thing is intermediate population densities. Sparsely populated areas don't matter because no one lives there. Densely populated areas are easier to service. Sprawling, intermediate areas, where lots of people live but the economics of providing service suck, are a big problem. https://www.census.gov/history...
> With IPv6 it'd be totally possible to move to a static default, you are path::to::ISP::customerNumber::MAC and it's yours forever and everything you do is linked by default.
Future cash is worth less than cash today-- because cash today can be invested and made into more cash.
If your investment returns less than an investment is supposed to return-- less than, say, 9% per annum-- then even if you've made "profit" in absolute terms you've made an extraordinarily poor return on capital (you'd have just been better off putting it in the market than going through all this crap of making a movie with it and taking various outsized risks on the way to make less money).
> I mean I can put a stack of $100 bills in the fireplace too and they will also be 'gone forever' as far as I am concerned personally. Its not like I can phone of the fed and ask them to print me some new ones.
He didn't? Read it more closely. He's highlighting the danger that the police's response to call like this poses.
I think everyone-- the people making the false call, and police's aggressive response to situations like this-- is responsible. I think it deserves a harsh penalty for the false reporters, because there's been such a pattern of behavior and such a flippant response after the death (the media interviews doubling down on swatting, etc, after his actions significantly contributed to someone's death are really something else)-- maybe not life in prison but a significant sentence.
> Even the most pessimistic figures we have today would suggest that self-driving cars are safer than the average driver.
Except they're comparing driving under the most favorable conditions (daylight, little snow or rain, mostly-highway, roads in good repair, etc)--- situations in which accidents are considerably rarer--- to an average across all humans at all times, including criminal behavior (deliberate recklessness, drunkenness, etc).
The US is a very large portion of Tesla's market. A subsidy shifts the demand curve, and Tesla both stands to move more units at better margin if it is able to deliver vehicles that qualify for it.
Have you considered a plug-in hybrid? I'm really liking my Clarity (I've burnt very little gas with the 45 mile electric range, but it's fully road trip capable).
Slashdot Mirror
Yah. There's good reasons why a minutes-long delay is not great, because we want things to settle out relatively quickly... but 2.5-25 milliseconds of fuzz on the order book would be just fine.
Why do you still insist it's physically impossible, given that lots of people used it and liked it 7 years ago on connections that were half the speed and double the latency of what's commonly available now? "Physically impossible" is an extraordinary claim. 500 miles at the speed of light is 2.5 milliseconds.
All I'm saying is I played it and I liked it, but games were going to be too uneconomical to justify it. That's pretty much what most of the reviews said, too. This seems to rule out "physically impossible" on the face, because a physical instance actually existed, when consumer internet connections had like one third the speed and double the last mile latency that they have today. Likewise, PS Now's current market success seems to rule out "physically impossible".
Also, you're a dipshit--- you can hardly shill for something that no longer exists. :P
Google... developed cert pinning (HPKP) and only after bad operational experience removed it:
https://www.zdnet.com/article/...
I agree it's fair, but it's not smart. If an employer is still interested in you and you've moved on, no need to alienate them by just ghosting-- if you say "thanks, but this isn't a fit" or "I've found something else"-- it means that whatever positive impression you've created with them can possibly still be useful to you in the future instead of creating a few people who feel the opposite.
That's just not how it went. And in any case, physical limits are hardly preventing PS Now from finding success and getting good reviews-- though people still say you're way better off with ethernet than wifi presently. Even that isn't a fundamental physical limit.
Onlive scoring good (not perfect) reviews for game performance in 2010-2011, as demonstrated below, with much worse than current consumer internet connections... is evidence that we're not up against physical limits. But it was an awful deal, and that + the Microsoft litigation killed them. After the fire sale of assets the service was indeed awful, from what I understand.
Top 3 reviews of found googling the words onlive 2011:
https://www.engadget.com/2010/...
> With an up-to-18Mbps AT&T U-Verse connection in San Jose, California, we found OnLive games loaded as quickly as on console -- sometimes much quicker -- and were actually quite playable. The controller never felt quite as responsive as that of a dedicated console nor the images quite as crisp, but we'd say that most of the time the overall experience was only slightly behind what we expect, only bogged down by the occasional annoying stutter. Frantic first-person shooters and driving games weren't as accurate as we like, but over the course of a couple days we adjusted to the mild lag, racking up plenty of kills, scoring the occasional headshot and drifting around some fairly tight corners as well. In Prince of Persia, a game that can require fairly precise timing in combat, we were still able to parry foes' swords and execute tricky jumps with a little bit of forethought, and a multiplayer game of Unreal Tournament III was intriguingly balanced -- if slightly laggy -- thanks to the fact that all players had 0 ping to the (virtual) host server.
https://www.pcgamer.com/onlive...
> And yet streaming from the net via OnLive is remarkably playable. Obviously it feels a bit sluggish compared with playing on your own native hardware, but for many games, especially those designed with laggy console controllers in mind, including the likes of Arkham Asylum and Human Revolution, it's far from unpleasant.
http://www.businessinsider.com...
> The game had minimal loading times, and while the graphics weren't as crystal clear as on a video game console (because of OnLive's compression technology), the level of detail was pretty amazing. It looks just as good as watching Netflix streaming. Controls originally felt a little delayed, but after a few minutes I felt right at home. I wanted to notice latency and laggy controls (due to my input getting beamed to the over the web, then a response getting beamed back), but I didn't find any in this game.
I'm not shilling for it. I bet at the time it would fail. It was an *awful* deal for games.
Even now, PS Now is doing alright, despite a much smaller deployed footprint (and worse user experience/average latency).
The biggest problems aren't the speed-of-light latency increase (physics) but things like other people in your house using the internet connection (bufferbloat/no reasonable QOS in most consumer edge routers).
Plenty of people play with TVs that add 100ms of latency. 20ms of round trip latency plus 10ms of encoder latency is not the end of the world.
I played Onlive for a month, it was reasonably good. The problems were A) massive capital burn from needing to run datacenters everywhere, and B) shitty deals with publishers, so the economics of buying on Online were awful.
It doesn't take *that* many datacenters to be within 20ms of 90% of the US population.
... Synthesis tools already exist, and every fab has a design library of standard transistors. While the tools are complicated and very expensive (though open source versions exist), they are there. So the problem you're describing is already solved. Designers describe logic, help a little with floorplan and constraints, and get a design out minutes to hours later.
What's difficult is that we don't have great programming mechanisms to describe parallel logic, or to synthesize sequential descriptions of tasks into efficient sets of gates and transistors. Verilog/VHDL/even SystemC are very, very, very slow and cumbersome to develop in and require a shit-ton of verification. If these could be made even a little more like developing software, it would be a big win.
But Onlive already did it, and it worked relatively well...
Or if you're talking about the wireless case, --- with present technology it's pretty difficult, but there's nothing fundamental to prevent it.
It's reasonable to expect that naughty code can be contained-- we have process containment and virtual machines and OS privileges for the purpose. But side channels, like Spectre etc, make that more difficult.
Luls premadonna
I've been talking about this a lot lately-- the refereeing in the NBA attracts a lot of complaints.
Right now there's just way too much stuff for referees to look for, at once. There are some things that computers could do very well (3 seconds in the key, lane violations, calling player/ball in/out of bounds, backcourt violation, midcourt 8 second timer). This would free up referees to look for traveling, off-ball contact, etc, so that the overall quality of officiating could improve.
The drivers testing the vehicles are employees. :P
Unless there's some kind of systemic bias --- in what kind of customers someone gets (e.g. ratings are poorer at night)--- it should average out and you should be able to use an average of different customers' types of ratings to compare employees. I agree yes or no questions, and long form "please explain" have value *also* in teasing apart what the customer experience is actually like.
You're missing the fact that 90-95% of humans have HHV6/HHV7, which is what the article is about, which are herpes viruses but not the cause of oral or genital herpes.
Hey, I'm a big Apple fan. The thing that's unfortunate right now is: if you are at a point where you should upgrade systems--- laptop life, OS support, etc--- all of the offerings are underwhelming: dated and not price-performant. Apple has always been a premium option but you'd usually get premium, up to date hardware for it until the past few years.
I'm on a Linux laptop these days and I hope they fix it so I can go back to everyday use of MacOS.
Your population density overall is lower because there's a lot of unpopulated land, but 75% of your population lives in areas with more than 1000 people per square kilometer :P.
The US has telecom problems beyond the population density one, but the bad thing is intermediate population densities. Sparsely populated areas don't matter because no one lives there. Densely populated areas are easier to service. Sprawling, intermediate areas, where lots of people live but the economics of providing service suck, are a big problem. https://www.census.gov/history...
> With IPv6 it'd be totally possible to move to a static default, you are path::to::ISP::customerNumber::MAC and it's yours forever and everything you do is linked by default.
RFC4941.
Future cash is worth less than cash today-- because cash today can be invested and made into more cash.
If your investment returns less than an investment is supposed to return-- less than, say, 9% per annum-- then even if you've made "profit" in absolute terms you've made an extraordinarily poor return on capital (you'd have just been better off putting it in the market than going through all this crap of making a movie with it and taking various outsized risks on the way to make less money).
Bullshit, as everyone knows.
If people weren't enslaved and killed over your engagement ring, it's just not true love.
> I mean I can put a stack of $100 bills in the fireplace too and they will also be 'gone forever' as far as I am concerned personally. Its not like I can phone of the fed and ask them to print me some new ones.
http://bep.gov/services/curren...
Though they note:
> Whoever mutilates currency with the intent to render it unfit to be reissued may be fined and/or imprisoned. 18 U.S.C. 333;
And so they'll probably both fine/imprison you and "print you some new ones".
> By advocating such a harsh sentence
He didn't? Read it more closely. He's highlighting the danger that the police's response to call like this poses.
I think everyone-- the people making the false call, and police's aggressive response to situations like this-- is responsible. I think it deserves a harsh penalty for the false reporters, because there's been such a pattern of behavior and such a flippant response after the death (the media interviews doubling down on swatting, etc, after his actions significantly contributed to someone's death are really something else)-- maybe not life in prison but a significant sentence.
> Even the most pessimistic figures we have today would suggest that self-driving cars are safer than the average driver.
Except they're comparing driving under the most favorable conditions (daylight, little snow or rain, mostly-highway, roads in good repair, etc)--- situations in which accidents are considerably rarer--- to an average across all humans at all times, including criminal behavior (deliberate recklessness, drunkenness, etc).
The US is a very large portion of Tesla's market. A subsidy shifts the demand curve, and Tesla both stands to move more units at better margin if it is able to deliver vehicles that qualify for it.
Have you considered a plug-in hybrid? I'm really liking my Clarity (I've burnt very little gas with the 45 mile electric range, but it's fully road trip capable).