(Warning, very handwaving explanation which most likely has technical inaccuracies but paints a basic picture.)
Basically, a ramjet relies on the inertia of an existing flow of air into its inlet to provide compression. That air continues to move to the back of the engine, and usually after the combustion chamber there is an expansion nozzle which allows the heated exhaust air to essentially push the engine forward with more force than it is resisting the incoming airflow.
As a result, for proper operation, a ramjet MUST have a source of airflow into the engine's inlet. This can be moving air into a stationary engine (practically, this means wind tunnel tests and no other situations), or forward inertia of the engine itself moving through stationary air. This is why unlike turbine-based jet engines, ramjets and scramjets have a minimum airspeed at which they will operate.
No 17" widescreen available, no Core Duo available, and most importantly, even the top end model carried by FIC (linked from your link) has integrated graphics.
If you want a cheapo POS laptop, yes, you can find one without Windows preinstalled/build most of it yourself.
Unfortunately, if you want a decent high-quality machine with a decent feature set, you still have to buy premade.
You can't blame Morton Thiokol (the contractor responsible for the SRBs) for the first Challenger incident.
They told NASA that the O-rings could fail below 53 degrees Fahrenheit and raised a HUGE stink about it prior to launch. No one at MT wanted to sign off on the launch, but the NASA administration forced them into it, essentially telling MT that they had to *prove* that there would be a problem, as opposed to NASA proving that ther could not be a problem.
FYI the Boeing 777 uses an onboard network that is Ethernet-based - Not sure how far they deviate from Ethernet spec, they may just be using uprated cabling and connectors. There are ARINC standards for Ethernet cabling and connectors on aircraft, the connectors specified are incredibly beefy and the cables are well shielded.
I also wonder whether SpaceX did their redundant backups Boeing style (define the interfaces, have independent hardware and software teams design and build each of the redundant boxes so the backups are different) or Ariane style (all boxes are identical, which has resulted in at least one Ariane launch failure because the bug that caused the primary flight control system to fail killed the backups within seconds of them coming online.)
Also, a good portion of that satellite's cost was probably NRE. Making a second one would probably be cheaper.
Although it's a student satellite, which drops NRE costs WAY down. (Students USUALLY work for free in exchange for course credit on projects like this. In fact, in many cases the students are effectively paying to work on the project.:)
Also, 3GPP doesn't cover all 3G phones. 3G GSM yes, 3G CDMA2000 no. CDMA2000 1xEV-DO is 3G and has no requirement for SIM cards, and in fact does not even support SIM cards.
Probably a small startup would fit into that category.
Larger companies which maintain strict control of company-owned machines and forbid bringing personal machines into the facility will often allow personal machines to connect to the company network via VPN for telecommuting purposes. The one restriction is that those machines (usually) need to be running some form of Windows.
Did you read what he said? Did you think about it at all?
Virtualization simply does not not provide acceptable for certain classes of applications. Anything involving 3D acceleration falls into this category. Note that the guy you replied to mentioned "games" in his post.
Also, reducing dependency hell with USE flags would be my second favorite feature. If I don't need support for certain things and said support is a pain in the ass to keep around, I can just make sure it isn't linked in.
"One size fits all" doesn't work for operating systems.
Stuff that works very well for certain types of users may be incredibly annoying for other types.
For example, Ubuntu is designed to be very friendly to new users. As a power user, the first time I tried Ubuntu it was like bashing my head against a brick wall repeatedly. It's a great distro for many people, it's just *not* for me. (And IMHO, not for anyone trying to set up a server machine.)
At the opposite end of the spectrum, Gentoo is an excellent distro for experienced power users, but it's a nightmare for new users. (In fact, it gives those new users more than enough rope to hang themselves. New Gentoo users typically push their CFLAG optimizations to insane and unstable limits because they can and it's cool, in the end breaking their system. Those are the users that the "Gentoo is for Ricers" page targets.) This is why I use Gentoo but would not ever reccommend it to a Linux newbie.
Fedora Core is somewhere in the middle ground between Ubuntu and Gentoo. As such, it tends to be the distro I reccommend to new people who want a system that's reasonably easy to get started with but still allows you to become familiar with the "down and dirty" details of a Linux installation.
Irrelevant to any of the phones covered by this article/discusssion. Both Treos and PocketPC platforms have no dependency on Get It Now!, and in fact don't even support Get It Now!.
System management stuff usually hangs off of an i2c or SMBus controller integrated into the motherboard chipset these days, not off the ISA bus.
As to the issue of changes in a point release - I suggest reading other posts describing changes made to Linux version numbering during the 2.6 series (which accompanied changes in the entire Linux development and release process.)
x.y.z versions can no longer be considered "point releases". That's why we now have an additional minor version number now (x.y.z.n). e.g. 2.6.15, 2.6.15.1, 2.6.15.2.
In short, version numbers are meaningless. And in this case, imagine that presence of an ISA bus in the system could potentially cripple performance of the PCI bus. There's a good chance that retaining support for the old dev=x,y,z system was causing problems. In fact, I'm pretty sure it WAS because it was giving the people writing code for removable devices (USB and Firewire storage) massive headaches. Interestingly enough, I recall that the first kernel version where hotpluggable USB and Firewire storage devices worked reliably was one of the ones you mention as breaking the old dev=x,y,z system.
"* Answering machine. Who needs voice mail on the provider side? Your phone probably has memory onboard + expansion slot memory. It has enough brains to record voice memos, do voice dialing, and play MP3s as ringtones. How hard is this to implement? Plus there's no monthly fee!" Voicemail is included as a part of every phone plan I've ever seen. Most importantly, having the answering machine in the phone is Just Plan Stupid. What if the phone is turned off or in an area where coverage is blocked? I would prefer to receive voicemails for incoming calls made when I'm at work, thank you very much.
"* Smart synchronization with Palm or WinCE PDAs. Most smart phones have bluetooth, but so far I have yet to find a way to sychronize the smartphone with the PDA in any useful way. Don't we have vcards and other standards for this?" Um, what crack are you on? The smartphone (in almost all cases) IS a PalmOS or WinCE PDA. Really, every smartphone I've seen (except maybe the Qualcomm pdQ) did a pretty good job of synchronizing with itself.:)
"* Smart synchronization with a PC. Even just a stupid Windows client + some documentation would be fine. I can write something that'll let my Linux desktop sync if it's documented! This could be as simple as dumping the data from the internal memory to the expansion memory in a parsable format, and then restoring it the same way -- the PC could have a program to read the memory card and deal with the data." If it syncs with a Palm or PocketPC device, it'll sync with a smartphone. I can use ANY sync software compatible with PalmOS to sync with my Treo 650. Pilot-link in Linux, Palm Desktop in Windows, any of the third-party Outlook conduits if I wanted to, Evolution in Linux in theory but Novell hasn't bothered to test sync functionality whatsoever since the Evo 1.x days. Evo just plain doesn't sync properly with any Palm device, every time I've tried it's been dupe city. jpilot works well though and I haven't tried Kontact yet.
"* Some kind of automation system. I have run across lots of little situations where I need to do something to a lot of contacts (move them into a group, delete duplicates, etc), and have found there's no batch interface. You have to deal with everything one click at a time."
Palm Desktop does that very well, and it's the default sync/management software for any PalmOS based smartphone. Don't know how the PocketPC PDAs fare in this regard.
Not sure about the privacy mode and such, I believe there is third party software for that (such as Profiles for PalmOS, which is VERY stable and free). Yes, it is third party software, but it isn't necessarily bugridden (such as Profiles).
I've never had any problems even remotely similar to that with my Verizon Treo 650.
As another poster suggested, TCPMP is GREAT. I use it on my Treo, and I've heard the PocketPC version (for the original poster) is also excellent.
Also in the category of "stuff you can do with a huge memory card", look into either Mapopolis or TomTom plus a Bluetooth GPS receiver. Like TCPMP, there are both PocketPC and PalmOS versions. TomTom seems to be getting more popular these days than Mapopolis.
Unfortunately, one of the nicest features of the early wireless PDAs and smartphones (web clipping applications) died when Palm shut down the Palm.net proxy servers. It's sad, because in terms of user interface, WCAs were one of the fastest ways of getting particular types of information. Using a web browser just isn't the same. Palm should have rewritten the WCA system to do normal HTML requests but still keep the WCA user interface. Note that there was an application recently released for PalmOS that seemed very similar to the original WCA system, I forget the name (look through the mytreo.net news archives if you use a Treo or other PalmOS smartphone), i'm not sure if there is a PocketPC equivalent.
To the parent of this post: Every single one of the apps listed above was installed to my Treo 650 without paying Verizon a single cent. Of course, with the exception of TCPMP and demo versions of the other apps, you will have to pay the original application developer some money. (Quite a bit in the case of Mapopolis or TomTom, thank Navtec's exorbitant licensing fees for that!)
I think his point is going back to the earlier posts that Schilling has been intentionally disabling any support in cdrecord except for the utterly horrific and user-unfriendly dev=x,y,z interface. i.e. he has intentionally broken it with respect to new kernels. The kernels have made progress, Schilling refuses to follow that progress. Before you start saying the kernel should have stuck with the old way of doing things for consistency - Do you think we should have stayed with ISA even though PCI was incompatible with many older peripherals? Do you think we should stay with CardBus in favor of moving to ExpressCard?
1) dev=x,y,z is simply user unfriently, when everything else in the system refers to an IDE device with/dev/something 2) dev=x,y,z prevents the use of udev for mapping devices to human-readable and sensible names like/dev/dvdrw 3) Related to 2) - A particular device cannot be guaranteed to have a consistent dev=x,y,z mapping. Removable (Firewire or USB2) DVD-R or CD-R drives are the best example of this. My Pioneer DVR-105 in an external enclosure would start at 1,0,0, then move to 2,0,0, then 3,0,0 every time it was unplugged and replugged. Meanwhile, its/dev/dvdrw mapping stayed the same with every reconnect.
If they were USB, I could understand a desire to lower the latency of the system. Some USB headsets may actually have unacceptable latency for gaming.
That said, these are pure analog headphones. From the looks of it, highly overpriced ones. Philips HN100 active noise cancelling headphones (good for killing fan noise) run around $60-70 retail, and you can find them for $15-20 INCLUDING S&H on eBay.
These headphones also don't have a built-in boom mic, just a shitty lapel mic that is completely seperate from the headphones. If they were comfortable, had a built-in boom mic, and were USB based, then they might be worth $99. As it is, they look like they aren't even worth as much as my old $30-at-Best-Buy Aiwas, let alone my HN100s.
It seems like quite a few. Brewing is definately a hobby with geek appeal. Chemical engineering, mechanical engineering (building equipment), and even electrical engineering (temperature control, etc) are all part of the process.:)
It may be extremely good in Britan, but in the U.S. it's just OK. Not bad, but not incredibly good either. I've tasted many brown ales that were much better, such as Ithaca Nut Brown.
It may be the "Guinness Effect" - From what I've heard, Guinness is much better in Ireland than it is on this side of the pond, most likely due to shipping, or differences in recipe for American tastes.
Actually, I think I recall seeing that the Guinness sold in the U.S. is brewed somewhere in the U.S. The brewery is owned and operated by the same people, but if they are using the same recipe as in Ireland and not tweaking it to take into account differences in the local water and other environmental effects, it won't be the same even if the recipe is.
If you avoid the mass produced crap, there are lots of excellent beers in the U.S. Even some of the larger breweries like Sam Adams and Sierra Nevada are pretty decent, although the best beers can usually only be found within a few miles of their brewery. For example, beer from Harvest Moon in New Brunswick, NJ can only be bought at the brewpub due to New Jersey laws regarding brewpubs and microbreweries. (Brewpubs can serve on premises but may not distribute, microbreweries are just the opposite. It's a weird and annoying law.) Wagner Valley Brewery (part of Wagner Vineyards in Lodi, NY) has an incredible doppelbock, but even though there is nothing in New York law that prevents Wagner from distributing, the only ways to get their beer seem to be going to the brewery or having it shipped directly from them.
If you live in the U.S., I strongly reccommend checking out your local breweries. Even if you dislike "beer" (as sold by the crap breweries), you may be pleasantly surprised at the local micros. For one thing, almost all of the major brews are pilsners, in my case I discovered that I just don't like pilsners no matter how well they are brewed. Your tastes may run towards other varieties, like wheat beer, stouts, or porters. Guinness is the most well known example of a stout, but it's an extremely bitter one that has been heavily hopped. Some other stouts have almost no hops at all and have a very different smooth and creamy taste, such as Ithaca Stout and Bar Harbor Cadillac Mountain Stout (from the Bar Harbor Brewing Company in Bar Harbor, ME. BH beers are not available anywhere other than in the area around Bar Harbor itself, so if you ever go hiking in Acadia National Park, stop at the three breweries on the island.)
I've seen plenty of cases of people switching away from GNOME to other DEs (including myself, I now use KDE even though I used to utterly hate it.) due to GNOME becoming crippleware in the name of "usability", is there anyone who has gone the other way due to the changes?
Between KDE and GNOME, GNOME has always (and still is) more polished and less buggy, but at this point GNOME is so crippled that given a choice between all of the missing functionality that was removed from GNOME and KDE's minor annoyances, I now prefer KDE.
Yonah supposedly also includes virtualization support. (It was supposed to, I haven't seen any confirmation either way whether or not it actually is in there. Anyone know if it would show in/proc/cpuinfo and what specific flags I should look for? I've got a T2500-based laptop in the other room.)
Also I have not seen any confirmation yet that the new Sossaman actuall supports multiple CPU packages per system. Intel's comparison page for the Xeons lists all of the Xeons I've tried as being MP capable, while the new Xeon LVs are only listed with a system type of "DP" - the die/package already HAS two processors though.
As to the issue of shared vs. independent cache between the cores - yes, it looks like while most third-party specs for the Yonah list it as "1M + 1M", it is actually 2M of unified cache for both cores according to Intel's docs/marketing for the Yonahs. The 1M+1M spec on some vendor pages (such as NewEgg) might be a holdover from the Athlon 64 X2 (which has independent L2 caches) and the Pentium D (which also appears to have independent L2 caches).
That unified cache between the cores could provide a significant performance boost that would make up for the unpleasant shared FSB.
Slashdot Mirror
(Warning, very handwaving explanation which most likely has technical inaccuracies but paints a basic picture.)
Basically, a ramjet relies on the inertia of an existing flow of air into its inlet to provide compression. That air continues to move to the back of the engine, and usually after the combustion chamber there is an expansion nozzle which allows the heated exhaust air to essentially push the engine forward with more force than it is resisting the incoming airflow.
As a result, for proper operation, a ramjet MUST have a source of airflow into the engine's inlet. This can be moving air into a stationary engine (practically, this means wind tunnel tests and no other situations), or forward inertia of the engine itself moving through stationary air. This is why unlike turbine-based jet engines, ramjets and scramjets have a minimum airspeed at which they will operate.
No 17" widescreen available, no Core Duo available, and most importantly, even the top end model carried by FIC (linked from your link) has integrated graphics.
If you want a cheapo POS laptop, yes, you can find one without Windows preinstalled/build most of it yourself.
Unfortunately, if you want a decent high-quality machine with a decent feature set, you still have to buy premade.
You can't blame Morton Thiokol (the contractor responsible for the SRBs) for the first Challenger incident.
They told NASA that the O-rings could fail below 53 degrees Fahrenheit and raised a HUGE stink about it prior to launch. No one at MT wanted to sign off on the launch, but the NASA administration forced them into it, essentially telling MT that they had to *prove* that there would be a problem, as opposed to NASA proving that ther could not be a problem.
FYI the Boeing 777 uses an onboard network that is Ethernet-based - Not sure how far they deviate from Ethernet spec, they may just be using uprated cabling and connectors. There are ARINC standards for Ethernet cabling and connectors on aircraft, the connectors specified are incredibly beefy and the cables are well shielded.
I also wonder whether SpaceX did their redundant backups Boeing style (define the interfaces, have independent hardware and software teams design and build each of the redundant boxes so the backups are different) or Ariane style (all boxes are identical, which has resulted in at least one Ariane launch failure because the bug that caused the primary flight control system to fail killed the backups within seconds of them coming online.)
Also, a good portion of that satellite's cost was probably NRE. Making a second one would probably be cheaper.
:)
Although it's a student satellite, which drops NRE costs WAY down. (Students USUALLY work for free in exchange for course credit on projects like this. In fact, in many cases the students are effectively paying to work on the project.
Also, 3GPP doesn't cover all 3G phones. 3G GSM yes, 3G CDMA2000 no. CDMA2000 1xEV-DO is 3G and has no requirement for SIM cards, and in fact does not even support SIM cards.
Probably a small startup would fit into that category.
Larger companies which maintain strict control of company-owned machines and forbid bringing personal machines into the facility will often allow personal machines to connect to the company network via VPN for telecommuting purposes. The one restriction is that those machines (usually) need to be running some form of Windows.
Did you read what he said? Did you think about it at all?
Virtualization simply does not not provide acceptable for certain classes of applications. Anything involving 3D acceleration falls into this category. Note that the guy you replied to mentioned "games" in his post.
That's one of my #1 favorite features.
Also, reducing dependency hell with USE flags would be my second favorite feature. If I don't need support for certain things and said support is a pain in the ass to keep around, I can just make sure it isn't linked in.
"One size fits all" doesn't work for operating systems.
Stuff that works very well for certain types of users may be incredibly annoying for other types.
For example, Ubuntu is designed to be very friendly to new users. As a power user, the first time I tried Ubuntu it was like bashing my head against a brick wall repeatedly. It's a great distro for many people, it's just *not* for me. (And IMHO, not for anyone trying to set up a server machine.)
At the opposite end of the spectrum, Gentoo is an excellent distro for experienced power users, but it's a nightmare for new users. (In fact, it gives those new users more than enough rope to hang themselves. New Gentoo users typically push their CFLAG optimizations to insane and unstable limits because they can and it's cool, in the end breaking their system. Those are the users that the "Gentoo is for Ricers" page targets.) This is why I use Gentoo but would not ever reccommend it to a Linux newbie.
Fedora Core is somewhere in the middle ground between Ubuntu and Gentoo. As such, it tends to be the distro I reccommend to new people who want a system that's reasonably easy to get started with but still allows you to become familiar with the "down and dirty" details of a Linux installation.
Irrelevant to any of the phones covered by this article/discusssion. Both Treos and PocketPC platforms have no dependency on Get It Now!, and in fact don't even support Get It Now!.
System management stuff usually hangs off of an i2c or SMBus controller integrated into the motherboard chipset these days, not off the ISA bus.
As to the issue of changes in a point release - I suggest reading other posts describing changes made to Linux version numbering during the 2.6 series (which accompanied changes in the entire Linux development and release process.)
x.y.z versions can no longer be considered "point releases". That's why we now have an additional minor version number now (x.y.z.n). e.g. 2.6.15, 2.6.15.1, 2.6.15.2.
In short, version numbers are meaningless. And in this case, imagine that presence of an ISA bus in the system could potentially cripple performance of the PCI bus. There's a good chance that retaining support for the old dev=x,y,z system was causing problems. In fact, I'm pretty sure it WAS because it was giving the people writing code for removable devices (USB and Firewire storage) massive headaches. Interestingly enough, I recall that the first kernel version where hotpluggable USB and Firewire storage devices worked reliably was one of the ones you mention as breaking the old dev=x,y,z system.
"* Answering machine. Who needs voice mail on the provider side? Your phone probably has memory onboard + expansion slot memory. It has enough brains to record voice memos, do voice dialing, and play MP3s as ringtones. How hard is this to implement? Plus there's no monthly fee!"
:)
Voicemail is included as a part of every phone plan I've ever seen. Most importantly, having the answering machine in the phone is Just Plan Stupid. What if the phone is turned off or in an area where coverage is blocked? I would prefer to receive voicemails for incoming calls made when I'm at work, thank you very much.
"* Smart synchronization with Palm or WinCE PDAs. Most smart phones have bluetooth, but so far I have yet to find a way to sychronize the smartphone with the PDA in any useful way. Don't we have vcards and other standards for this?"
Um, what crack are you on? The smartphone (in almost all cases) IS a PalmOS or WinCE PDA. Really, every smartphone I've seen (except maybe the Qualcomm pdQ) did a pretty good job of synchronizing with itself.
"* Smart synchronization with a PC. Even just a stupid Windows client + some documentation would be fine. I can write something that'll let my Linux desktop sync if it's documented! This could be as simple as dumping the data from the internal memory to the expansion memory in a parsable format, and then restoring it the same way -- the PC could have a program to read the memory card and deal with the data."
If it syncs with a Palm or PocketPC device, it'll sync with a smartphone. I can use ANY sync software compatible with PalmOS to sync with my Treo 650. Pilot-link in Linux, Palm Desktop in Windows, any of the third-party Outlook conduits if I wanted to, Evolution in Linux in theory but Novell hasn't bothered to test sync functionality whatsoever since the Evo 1.x days. Evo just plain doesn't sync properly with any Palm device, every time I've tried it's been dupe city. jpilot works well though and I haven't tried Kontact yet.
"* Some kind of automation system. I have run across lots of little situations where I need to do something to a lot of contacts (move them into a group, delete duplicates, etc), and have found there's no batch interface. You have to deal with everything one click at a time."
Palm Desktop does that very well, and it's the default sync/management software for any PalmOS based smartphone. Don't know how the PocketPC PDAs fare in this regard.
Not sure about the privacy mode and such, I believe there is third party software for that (such as Profiles for PalmOS, which is VERY stable and free). Yes, it is third party software, but it isn't necessarily bugridden (such as Profiles).
This doesn't help you, but for any readers running PalmOS based smartphones, pssh ROCKS.
I've never had any problems even remotely similar to that with my Verizon Treo 650.
As another poster suggested, TCPMP is GREAT. I use it on my Treo, and I've heard the PocketPC version (for the original poster) is also excellent.
Also in the category of "stuff you can do with a huge memory card", look into either Mapopolis or TomTom plus a Bluetooth GPS receiver. Like TCPMP, there are both PocketPC and PalmOS versions. TomTom seems to be getting more popular these days than Mapopolis.
Unfortunately, one of the nicest features of the early wireless PDAs and smartphones (web clipping applications) died when Palm shut down the Palm.net proxy servers. It's sad, because in terms of user interface, WCAs were one of the fastest ways of getting particular types of information. Using a web browser just isn't the same. Palm should have rewritten the WCA system to do normal HTML requests but still keep the WCA user interface. Note that there was an application recently released for PalmOS that seemed very similar to the original WCA system, I forget the name (look through the mytreo.net news archives if you use a Treo or other PalmOS smartphone), i'm not sure if there is a PocketPC equivalent.
To the parent of this post: Every single one of the apps listed above was installed to my Treo 650 without paying Verizon a single cent. Of course, with the exception of TCPMP and demo versions of the other apps, you will have to pay the original application developer some money. (Quite a bit in the case of Mapopolis or TomTom, thank Navtec's exorbitant licensing fees for that!)
I think his point is going back to the earlier posts that Schilling has been intentionally disabling any support in cdrecord except for the utterly horrific and user-unfriendly dev=x,y,z interface. i.e. he has intentionally broken it with respect to new kernels. The kernels have made progress, Schilling refuses to follow that progress. Before you start saying the kernel should have stuck with the old way of doing things for consistency - Do you think we should have stayed with ISA even though PCI was incompatible with many older peripherals? Do you think we should stay with CardBus in favor of moving to ExpressCard?
/dev/something /dev/dvdrw /dev/dvdrw mapping stayed the same with every reconnect.
1) dev=x,y,z is simply user unfriently, when everything else in the system refers to an IDE device with
2) dev=x,y,z prevents the use of udev for mapping devices to human-readable and sensible names like
3) Related to 2) - A particular device cannot be guaranteed to have a consistent dev=x,y,z mapping. Removable (Firewire or USB2) DVD-R or CD-R drives are the best example of this. My Pioneer DVR-105 in an external enclosure would start at 1,0,0, then move to 2,0,0, then 3,0,0 every time it was unplugged and replugged. Meanwhile, its
If they were USB, I could understand a desire to lower the latency of the system. Some USB headsets may actually have unacceptable latency for gaming.
That said, these are pure analog headphones. From the looks of it, highly overpriced ones. Philips HN100 active noise cancelling headphones (good for killing fan noise) run around $60-70 retail, and you can find them for $15-20 INCLUDING S&H on eBay.
These headphones also don't have a built-in boom mic, just a shitty lapel mic that is completely seperate from the headphones. If they were comfortable, had a built-in boom mic, and were USB based, then they might be worth $99. As it is, they look like they aren't even worth as much as my old $30-at-Best-Buy Aiwas, let alone my HN100s.
It seems like quite a few. Brewing is definately a hobby with geek appeal. Chemical engineering, mechanical engineering (building equipment), and even electrical engineering (temperature control, etc) are all part of the process. :)
It may be extremely good in Britan, but in the U.S. it's just OK. Not bad, but not incredibly good either. I've tasted many brown ales that were much better, such as Ithaca Nut Brown.
It may be the "Guinness Effect" - From what I've heard, Guinness is much better in Ireland than it is on this side of the pond, most likely due to shipping, or differences in recipe for American tastes.
Actually, I think I recall seeing that the Guinness sold in the U.S. is brewed somewhere in the U.S. The brewery is owned and operated by the same people, but if they are using the same recipe as in Ireland and not tweaking it to take into account differences in the local water and other environmental effects, it won't be the same even if the recipe is.
If you avoid the mass produced crap, there are lots of excellent beers in the U.S. Even some of the larger breweries like Sam Adams and Sierra Nevada are pretty decent, although the best beers can usually only be found within a few miles of their brewery. For example, beer from Harvest Moon in New Brunswick, NJ can only be bought at the brewpub due to New Jersey laws regarding brewpubs and microbreweries. (Brewpubs can serve on premises but may not distribute, microbreweries are just the opposite. It's a weird and annoying law.) Wagner Valley Brewery (part of Wagner Vineyards in Lodi, NY) has an incredible doppelbock, but even though there is nothing in New York law that prevents Wagner from distributing, the only ways to get their beer seem to be going to the brewery or having it shipped directly from them.
If you live in the U.S., I strongly reccommend checking out your local breweries. Even if you dislike "beer" (as sold by the crap breweries), you may be pleasantly surprised at the local micros. For one thing, almost all of the major brews are pilsners, in my case I discovered that I just don't like pilsners no matter how well they are brewed. Your tastes may run towards other varieties, like wheat beer, stouts, or porters. Guinness is the most well known example of a stout, but it's an extremely bitter one that has been heavily hopped. Some other stouts have almost no hops at all and have a very different smooth and creamy taste, such as Ithaca Stout and Bar Harbor Cadillac Mountain Stout (from the Bar Harbor Brewing Company in Bar Harbor, ME. BH beers are not available anywhere other than in the area around Bar Harbor itself, so if you ever go hiking in Acadia National Park, stop at the three breweries on the island.)
The first time I saw this post was in an article on Enigma.
On the other hand, from what I've read of Enigma, that guy's post isn't Enigma-encrypted. I could be wrong though.
It's probably an encrypted GNAA troll.
I've seen plenty of cases of people switching away from GNOME to other DEs (including myself, I now use KDE even though I used to utterly hate it.) due to GNOME becoming crippleware in the name of "usability", is there anyone who has gone the other way due to the changes?
Between KDE and GNOME, GNOME has always (and still is) more polished and less buggy, but at this point GNOME is so crippled that given a choice between all of the missing functionality that was removed from GNOME and KDE's minor annoyances, I now prefer KDE.
The only Ambilight model listed on Philips' website is a 42" LCD unit, not a plasma.
Yonah supposedly also includes virtualization support. (It was supposed to, I haven't seen any confirmation either way whether or not it actually is in there. Anyone know if it would show in /proc/cpuinfo and what specific flags I should look for? I've got a T2500-based laptop in the other room.)
Also I have not seen any confirmation yet that the new Sossaman actuall supports multiple CPU packages per system. Intel's comparison page for the Xeons lists all of the Xeons I've tried as being MP capable, while the new Xeon LVs are only listed with a system type of "DP" - the die/package already HAS two processors though.
As to the issue of shared vs. independent cache between the cores - yes, it looks like while most third-party specs for the Yonah list it as "1M + 1M", it is actually 2M of unified cache for both cores according to Intel's docs/marketing for the Yonahs. The 1M+1M spec on some vendor pages (such as NewEgg) might be a holdover from the Athlon 64 X2 (which has independent L2 caches) and the Pentium D (which also appears to have independent L2 caches).
That unified cache between the cores could provide a significant performance boost that would make up for the unpleasant shared FSB.
The article specifies a TDP of 31W, a total of 2 MB cache (1M per core), 667 MHz FSB, and a clock speed of 2.0 GHz.
How is this different from the Core Duo T2500? From the looks of it, there is none.
Gentoo 2005.1 installed with no problems on the SATA-based Dell PowerEdge 850 (or was it 830???) I set up for him.
Same with my SATA-based Athlon X2 server.
Gentoo 2006.0 was flawless on my new Inspiron E1705.