Ahem, high definition, anyone? At a raw bitrate of over 120 MB/s, and _maybe_ halfway decent compressed quality at 8 - 12 MB/s (personally I'd say not but I spent a summer looking at the real thing), bigger optical media are a prerequisite for high definition.
Otherwise, you're looking at maybe 20 MINUTES of footage on a disk, max.
Yep, I don't know why I thought it was a 6502, you are quite right - looking at it, the Megadrive has a Z80A controlling the sound playback, with both a Yamaha YM2612 6 channel FM synthesis chip and a TI 76489 4 channel sound generator.
even the dookie sound chip in the MD/Genesis could be refabbed with an interface that liberates it and affords it a special niche all its own.
Perhaps not; wasn't the sound on the Genesis/Megadrive handled by a separate 6502 8-bit CPU acting as a copro to the main 68000 CPU, rather than custom hardware?
Thinking about it, did that 6502 run some sort of audio soft synthesis code, or was it completly up to the game designers to program it to generate sounds?
Sort of; there are two concepts here, the total amount of energy stored (found from the capacity, expressed in Ah, multiplied by the voltage) and the rate you can get that energy in or out (confusingly given as a multiple of the capacity even though that makes no sense dimensionally). For example, a 10 Ah NiCd battery can be safely discharged at a rate of 2C (20 A) for 30 minutes (10 A for 1 hour == 20A for 0.5 hour).
To answer your question, a little battery contains enough energy to power that computer, but you can't get the energy out fast enough (safely) to do so.
The physical construction and chemistry of the battery determines the maximum safe charge / discharge rates, for example sintered plate NiCd cells can supply more current than mass plate designs of the same physical size.
Yes, but do you know WHY? Lets do the maths: say a rechargable D-sized cell produces 2.0V (chemistry, no way round that) and has a capacity of 2.5Ah. That energy translates to an optimistic maximum of about 5 Watts power for about an hour. Note that there is a maximum safe discharge rate - the cell can easily supply 400 A into a short circuit, and self-destruct in the process. Draw more amps than is safe and the cell may produce hydrogen and oxygen (in a closed space, too, yay!). Often the maximum rate is about 3 times the capacity (3C) for lead-acid, so we can safely draw 7.5A for 20 minutes, given 15W.
Compare that with the 90-120 Watts design power of your CPU. Add in RAM, disks, motherboard, PCI cards (assuming you can't get these to power down cleanly, fast, WITHOUT losing disk access; no point powering the disks if the RAID card just went away), and you'll probably need 200W.
Now you need to get the right voltages out: 1.5V core, 3.3V & 5V, 12V for disks, -5V for RAM. That's a lot of converters, and some at high power, implying switching regulators, heat sinks, maybe forced cooling (ooops, that's more power needed for the fans) since you want this internal. Assume 10-15% losses, so now you need 230W. That's more than FIFTEEN D-sized cells for your "little" UPS. On the plus side, it'll run for 20 minutes, not 2 minutes.
Upshot is, lead acid battery tech. isn't going to supply 200W for 2 minutes in a reasonable size, because the safe discharge rate limits the power.
For comparison, lithium polymer cells are limited to usually 5C rates, but can be as high as 20C - now those might work in this application, but now you're talking hundreds of dollars JUST for the battery.
I seem to remember a study that said otherwise. Do you have a source?
I thought it was Honeynet but that is quite old, and only talks about RH6.2. Do you have a source for a more up to date study?
Thanks for the info on other distributions. FYI, we use Debian primarily for intranet servers but occasionally for day-to-day development. I'm a developer not a sysadmin, so I don't have the time to stay up to date on anything we don't actually run. We looked at RHEL primarily for the support but were unimpressed to the point of standardising on Debian.
You only see the 1/4 memory space IE uses because the rest of it is incorporated into the OS. If core Firefox components were incorporated into Windows the way IE is, it would show less memory usage than IE does.
Yeah, yeah, yeah. We keep hearing about this and it's bollocks. The HTML renderer is in mshtml.dll, and iexplore.exe is just a shell which wraps it. So what? It's not "incorporated into the OS", kernel32.dll & ntdll.dll don't link to it, and if you try you can unincoporate it. Explorer.exe links to it, meaning that the HTML renderer is loaded and running as part of the windowing shell, but guess what, you can kill the shell and start programs manually using task manager, or even use a different shell. Hell, if you REALLY care that fucking much, get an XP Embedded license and build your own XP, with EXACTLY the components you want. Should keep you busy for hours, just like a Gentoo build from scratch.
I bet Firefox doesn't even dynamically rebase its DLLs to optimise load times (hint, ActiveState Perl does this on installation, and almost all MS apps do this as a matter of course) - saves the PE loader from doing it for you at application load time.
But I can put up a linux box of any flavor, attach it to the network, lock it down in minutes, and not worry about it being compromised.
Oh really? I seem to remember an investigation which concluded that the lifetime of an unpatched box on a public network was about 5 minutes, irregardless of OS. When was that last OpenSSH flaw fixed? Has your vendor bothered to backport that fix? I might just believe you if you said "Debian" box, because the Debian people I know put a lot of work into fixing and backporting fixes, which apt is configured to pick up by default. I flat out do not believe that Redhat / Fedora ships secure, although at least RHEL seems to have some sort of maintained updates server. (and I don't use SUSE or Slackware so I have no idea about them).
Anyone connecting ANYTHING to the public network without a good firewall in the way first is asking for it, IME.
Apparently EDS attempted to do a test upgrade on a small network of 7 machines, but accidentally deployed it to all 80,000 machines instead. It's not clear that they'd tried it on any target machine, so it's entirely possible that EDS is to blame here...
Eclipse has CVS problems because they write their own client and now have to keep it up to date and bug compatible with the various CVS forks (yes, Tony, I mean CVSNT!).
CVS apparently is still not featureful enough to not have to rely on parsing the text output of various commands, and not stable enough that that text format doesn't keep changing (like we just discovered, Debian (testing) runs CVS 1.12 and Eclipse 3 only officialy supports CVS 1.11 due to some textual output change, WTF???).
Subversion support for Eclipse is a joke, Perforce is too expensive, Bitkeeper support is nonexistent... We are stuck with CVS.
One area I would like to see explored more is divorcing presentation from meaning (DPFM).
Yes, I couldn't agree more. On a similar theme, I'd like to see source control tools which store the parse tree of the code as the canonical source, rather than the ASCII (or Unicode, in the case of Java) source code. Then the local system regenerates the source on the users machine, with the correct local coding conventions, whitespace, indenting etc. No more "braces-go-here" wars, and tools which already use the parse tree (like Eclipse) have one fewer task to do on checkout of a large project.
Partly the problem is that C++ is a pain in the ass to parse - for example you need whitespace to stop nested templates > from mis-parsing as the stream operator >>, and if there's a syntactical error it's harder to recover gracefully and carry on parsing. This makes building fast parse trees in memory and maintaining them as the user keeps on typing harder than, say Java (notice how generics in Java 1.5 correctly parses > without needing magic whitespace?
Hmmm. We've used every milestone in the 3.0 series (except one) on Windows XP and Debian (sid) boxes, and not had problems. The Linux GTK version is even fast enough over remote X sessions to a Linux server using the cygwin X-server.
You do know that Eclipse uses SWT, not Swing, and SWT uses native widgets (so GTK or Motif on Linux)?
Admittedly all the extra features in Eclipse 3 have made it slower than 2 (and I almost prefer the v2 UI, it's cleaner and faster), but nothing is free...
I believe Eclipse will become a major part of software tools of the future, simply because the Rich Client Platform initiative brought in for Eclipse 3.0. This makes it easier to create custom Eclipse API applications, compared to version 2 which was a much less modular design.
My employer is working on an Eclipse-based development environment for their Cascade product, instead of developing yet another custom environment from scratch with all the incompatibility and test overhead that that entails.
I have had some very bad experiences with antivirus software that hooks the 16 bit subsystem (NTVDM processes). We're not talking no-brand crap here either.
You don't need AV software to know that your machines aren't, for example, spamming the world. I have a Linux mail server and I watch the goddamn logs.
You don't need AV software to "protect" Internet Explorer; I have a Linux server running a filtering proxy which amongst other things kills Javascript at source, and an administrative lockdown on ActiveX controls.
It's all bullshit, driven by people who want to sell you AV software. But then, I work for a small company where everyone has a clue.
Jon.
Re:Time to be a troll
on
RT Linux Patches
·
· Score: 2, Interesting
Ahem.
Real time linux is NOT "a true RTOS running linux as its lowest priority thread." That doesn't even make sense! You've obviously never done any kernel programming, or bothered to do any basic knowledge gathering on operating system design at all.
Perhaps you might like to tell certain large Japanese A/V companies about that, since they do EXACTLY that (run Linux as a task under a uITRON RTOS). I think you maybe need to learn something, too.
Just another reason why I won't run AV software on the desktop. _I_ don't need it, and my systems and networks are actually _more_ stable without it. None of my machines, nor the networks that I administer, have EVER had a virus.
Ditto for tubes. You can mangle the signal in software nowadays to simulate any kind of electronic or accoustic circuit or environment.
Up to a point. As a guitarist, I have not yet seen such a device which converts my 100W, sterile, accurate MOSFET amp into a valve amp. As an engineer, I'm sure it's possible, but it's a LOT harder than just messing with EQ. Even the modelling preamps (which seem to be convolving the sound with the FIR response of whatever they're modelling) don't get the ringing clarity and touch-sensitive response.
You're entitled to your opinion, but the reality is that valves still sound better than solid state to a lot of people.
That's an admirable superiority complex you have going there.
Er, no, that's the logical conclusion from the facts. Either you, or your admin, has re-enabled something which you consider to be a security problem. This isn't WinXP's problem! I didn't say I've never done anything similar;-)
I didn't make this corporate distro. It's enabled. We checked other machines around the office.
Then it's a domain config problem and your domain admin staff are at fault. Active Directory and Group Policy need to be configured correctly. Running a windows domain is NOT just a case of slapping together a server and creating user accounts. A default installation of a Windows 2000 server (I don't know about 2003 server) is to set all policy items as unconfigured, which is probably not what you want. There are lots of books, articles and MS deployment guides on configuring group policy correctly, and you can configure separate policies for the domain controllers and domain members, and associate different policies with security groups and LDAP Organisational Units in the directory.
You, and your domain admins, don't know what you are doing. First, RDC is disabled by default in Windows XP. Second, RDC can be locked down using group policy. Read this information.
It asks you which of the current users should be setup as a local system admin
This is not how you should set up a domain, by the way. There shouldn't BE any local users other than the local administrator. Domain user accounts are managed from the domain controller.
Usually we only have one user on each machine, and so they get the admin rights locally on it.
And why are you expecting this to be secure? Do you give everyone root on their own linux boxes as well? Any domain admin with a clue sets things up so that the domain users are "User" or "Power User" at best, and a lot of places lock things down even further using group policy. You can reduce XP to kiosk mode if necessary. I've actually deployed SP2 in a domain and the XP firewall can be configured using domain policy such that local admin can't mess with it.
Many Windows 2000 domains have a Windows DNS server to integrate with Active Directory, because older versions of Bind don't support the various features. See this tip for details.
Slashdot Mirror
Otherwise, you're looking at maybe 20 MINUTES of footage on a disk, max.
Jon.
Jon
Perhaps not; wasn't the sound on the Genesis/Megadrive handled by a separate 6502 8-bit CPU acting as a copro to the main 68000 CPU, rather than custom hardware?
Thinking about it, did that 6502 run some sort of audio soft synthesis code, or was it completly up to the game designers to program it to generate sounds?
Jon.
To answer your question, a little battery contains enough energy to power that computer, but you can't get the energy out fast enough (safely) to do so.
The physical construction and chemistry of the battery determines the maximum safe charge / discharge rates, for example sintered plate NiCd cells can supply more current than mass plate designs of the same physical size.
Jon.
Yes, but do you know WHY? Lets do the maths: say a rechargable D-sized cell produces 2.0V (chemistry, no way round that) and has a capacity of 2.5Ah. That energy translates to an optimistic maximum of about 5 Watts power for about an hour. Note that there is a maximum safe discharge rate - the cell can easily supply 400 A into a short circuit, and self-destruct in the process. Draw more amps than is safe and the cell may produce hydrogen and oxygen (in a closed space, too, yay!). Often the maximum rate is about 3 times the capacity (3C) for lead-acid, so we can safely draw 7.5A for 20 minutes, given 15W.
Compare that with the 90-120 Watts design power of your CPU. Add in RAM, disks, motherboard, PCI cards (assuming you can't get these to power down cleanly, fast, WITHOUT losing disk access; no point powering the disks if the RAID card just went away), and you'll probably need 200W.
Now you need to get the right voltages out: 1.5V core, 3.3V & 5V, 12V for disks, -5V for RAM. That's a lot of converters, and some at high power, implying switching regulators, heat sinks, maybe forced cooling (ooops, that's more power needed for the fans) since you want this internal. Assume 10-15% losses, so now you need 230W. That's more than FIFTEEN D-sized cells for your "little" UPS. On the plus side, it'll run for 20 minutes, not 2 minutes.
Upshot is, lead acid battery tech. isn't going to supply 200W for 2 minutes in a reasonable size, because the safe discharge rate limits the power.
For comparison, lithium polymer cells are limited to usually 5C rates, but can be as high as 20C - now those might work in this application, but now you're talking hundreds of dollars JUST for the battery.
Jon.
I thought it was Honeynet but that is quite old, and only talks about RH6.2. Do you have a source for a more up to date study?
Thanks for the info on other distributions. FYI, we use Debian primarily for intranet servers but occasionally for day-to-day development. I'm a developer not a sysadmin, so I don't have the time to stay up to date on anything we don't actually run. We looked at RHEL primarily for the support but were unimpressed to the point of standardising on Debian.
Jon.
And what do you think SUS Server (now called Windows Update Server, apparently) is for then? More blinkenlights for the collection?
Jon.
Yeah, yeah, yeah. We keep hearing about this and it's bollocks. The HTML renderer is in mshtml.dll, and iexplore.exe is just a shell which wraps it. So what? It's not "incorporated into the OS", kernel32.dll & ntdll.dll don't link to it, and if you try you can unincoporate it. Explorer.exe links to it, meaning that the HTML renderer is loaded and running as part of the windowing shell, but guess what, you can kill the shell and start programs manually using task manager, or even use a different shell. Hell, if you REALLY care that fucking much, get an XP Embedded license and build your own XP, with EXACTLY the components you want. Should keep you busy for hours, just like a Gentoo build from scratch.
I bet Firefox doesn't even dynamically rebase its DLLs to optimise load times (hint, ActiveState Perl does this on installation, and almost all MS apps do this as a matter of course) - saves the PE loader from doing it for you at application load time.
But I can put up a linux box of any flavor, attach it to the network, lock it down in minutes, and not worry about it being compromised.
Oh really? I seem to remember an investigation which concluded that the lifetime of an unpatched box on a public network was about 5 minutes, irregardless of OS. When was that last OpenSSH flaw fixed? Has your vendor bothered to backport that fix? I might just believe you if you said "Debian" box, because the Debian people I know put a lot of work into fixing and backporting fixes, which apt is configured to pick up by default. I flat out do not believe that Redhat / Fedora ships secure, although at least RHEL seems to have some sort of maintained updates server. (and I don't use SUSE or Slackware so I have no idea about them).
Anyone connecting ANYTHING to the public network without a good firewall in the way first is asking for it, IME.
Jon.
Jon.
CVS apparently is still not featureful enough to not have to rely on parsing the text output of various commands, and not stable enough that that text format doesn't keep changing (like we just discovered, Debian (testing) runs CVS 1.12 and Eclipse 3 only officialy supports CVS 1.11 due to some textual output change, WTF???).
Subversion support for Eclipse is a joke, Perforce is too expensive, Bitkeeper support is nonexistent... We are stuck with CVS.
Jon.
Yes, I couldn't agree more. On a similar theme, I'd like to see source control tools which store the parse tree of the code as the canonical source, rather than the ASCII (or Unicode, in the case of Java) source code. Then the local system regenerates the source on the users machine, with the correct local coding conventions, whitespace, indenting etc. No more "braces-go-here" wars, and tools which already use the parse tree (like Eclipse) have one fewer task to do on checkout of a large project.
Jon.
Partly the problem is that C++ is a pain in the ass to parse - for example you need whitespace to stop nested templates > from mis-parsing as the stream operator >>, and if there's a syntactical error it's harder to recover gracefully and carry on parsing. This makes building fast parse trees in memory and maintaining them as the user keeps on typing harder than, say Java (notice how generics in Java 1.5 correctly parses > without needing magic whitespace?
Jon.
You do know that Eclipse uses SWT, not Swing, and SWT uses native widgets (so GTK or Motif on Linux)?
Admittedly all the extra features in Eclipse 3 have made it slower than 2 (and I almost prefer the v2 UI, it's cleaner and faster), but nothing is free...
Jon.
My employer is working on an Eclipse-based development environment for their Cascade product, instead of developing yet another custom environment from scratch with all the incompatibility and test overhead that that entails.
Jon.
Man, that is one of the funniest song parodies I've seen posted. Is it original?
I have had some very bad experiences with antivirus software that hooks the 16 bit subsystem (NTVDM processes). We're not talking no-brand crap here either.
You don't need AV software to know that your machines aren't, for example, spamming the world. I have a Linux mail server and I watch the goddamn logs.
You don't need AV software to "protect" Internet Explorer; I have a Linux server running a filtering proxy which amongst other things kills Javascript at source, and an administrative lockdown on ActiveX controls.
It's all bullshit, driven by people who want to sell you AV software. But then, I work for a small company where everyone has a clue.
Jon.
Real time linux is NOT "a true RTOS running linux as its lowest priority thread." That doesn't even make sense! You've obviously never done any kernel programming, or bothered to do any basic knowledge gathering on operating system design at all.
Perhaps you might like to tell certain large Japanese A/V companies about that, since they do EXACTLY that (run Linux as a task under a uITRON RTOS). I think you maybe need to learn something, too.
Jon.
Just another reason why I won't run AV software on the desktop. _I_ don't need it, and my systems and networks are actually _more_ stable without it. None of my machines, nor the networks that I administer, have EVER had a virus.
Jon.
Services for UNIX is free; get it here.
Jon.
Up to a point. As a guitarist, I have not yet seen such a device which converts my 100W, sterile, accurate MOSFET amp into a valve amp. As an engineer, I'm sure it's possible, but it's a LOT harder than just messing with EQ. Even the modelling preamps (which seem to be convolving the sound with the FIR response of whatever they're modelling) don't get the ringing clarity and touch-sensitive response.
You're entitled to your opinion, but the reality is that valves still sound better than solid state to a lot of people.
Jon.
Er, no, that's the logical conclusion from the facts. Either you, or your admin, has re-enabled something which you consider to be a security problem. This isn't WinXP's problem! I didn't say I've never done anything similar ;-)
I didn't make this corporate distro. It's enabled. We checked other machines around the office.
Then it's a domain config problem and your domain admin staff are at fault. Active Directory and Group Policy need to be configured correctly. Running a windows domain is NOT just a case of slapping together a server and creating user accounts. A default installation of a Windows 2000 server (I don't know about 2003 server) is to set all policy items as unconfigured, which is probably not what you want. There are lots of books, articles and MS deployment guides on configuring group policy correctly, and you can configure separate policies for the domain controllers and domain members, and associate different policies with security groups and LDAP Organisational Units in the directory.
Jon.
Jon.
This is not how you should set up a domain, by the way. There shouldn't BE any local users other than the local administrator. Domain user accounts are managed from the domain controller.
Usually we only have one user on each machine, and so they get the admin rights locally on it.
And why are you expecting this to be secure? Do you give everyone root on their own linux boxes as well? Any domain admin with a clue sets things up so that the domain users are "User" or "Power User" at best, and a lot of places lock things down even further using group policy. You can reduce XP to kiosk mode if necessary. I've actually deployed SP2 in a domain and the XP firewall can be configured using domain policy such that local admin can't mess with it.
Jon.
Jon.
Jon.