I recently upgraded a friends PC from ME to XP Home. She purchased XP, which came with a sticker proclaiming that it included SP1a.
Since this was a recent purchase and the after thought SP1a sticker was there, I mistakenly assumed that it would be safe against Blaster.
Regardless, I enabled the built in firewall on the external interface NIC before I connected to the internet via her ADSL.
I couldn't get it going. I was using the ISP PPPoE driver which was supposed to work, but the ISP suggested I use the built in XP PPPoE driver, which worked fine. The phone tech also said that I must disable any firewall due to the use of a heartbeat initiated at their end.
So, I reluctantly did...
Her PC had Blaster literally within a minute or two of connecting.
But here comes the funny part... to get around the 60 seconds to shutdown, I double clicked the time to set the year back to give me a chance to remove the virus and patch her system. Unfortunately, during this, I had to reboot. At this stage the 30 day registration period was still in effect because I had not registered. Upon reboot, the 30 day period was up, XP was demanding I register now without giving me the desktop! Luckily it seems that it automatically connected.
Next time I'll just set it back an hour!
This kind of crap just has not happened to me on my Apple. In the end, I enabled the firewall and she has not had a problem. It might not have happened if I knew XP better (first install), but then I gave up on Microsoft long ago.
About 2 years ago I projected some images I took with some Agfa Scala 200 ASA film. Measured the size of the frame and the number of grains within a small square within that frame.
On average (the grains are random shapes and sizes, within limits), it came out to about 12 million grains within the full frame.
So it would take about 12 Mega pixels to match Scala 200 (maybe a little less than 12, to take into account the random nature of the grains which can make an image look less detailed due to the larger grains spoiling the effect of the whole). I imagine some of Fuji's highest resolution films would be 25 million grains or beyond.
I think the biggest hurdle to digital photography is not the CCD resolutions, but the available memory. Ever manipulated a high quality jpeg? The artifacts that were not noticable suddenly jump out at you when you run a filter, for example. So really high quality compression or lossless compression would be nice. A raw image at 25 Mpixels is a full 71 Megabytes for 24bit colour.
Choose a make who builds great lenses. Nikon, Canon, Carl Zeis. Buy good lenses from that company (read reviews which do proper testing for resolution, distortion, contrast, etc, not just some subjective crap written by some self proclaimed knowledgable photographer) and then find a body to go with them. Buying expensive bodies and cheap lenses is a very bad decision. Spending $20,000 on lenses and $200 on a body is not a bad thing to do. The effects of the body with a good photographer are truely negligible as long as the shutter speeds are accurate.
If you really want to learn photography, get a fully manual camera with a built in light meter (to learn the flaws of built in light meters) and use either grey cards or a hand held light meter.
Like the Pentax K-1000, it is built like a tank. Nikon takes two blocks of aluminium and machine the body and prism housing out of them. Very strong, 100% mechanical (besides the light meter). The AS variant is the best of the F2's which has a light meter, due to the sensor being silicon as opposed to the failure prone light dependant resistors of the previous models. The camera was released in 1977 and yet many current Nikkor lenses (and almost all Nikkor lenses) can be used with it.
I've been to places with friends where we've all tried to take a sudden "good shot", I've just had to raise my camera, focus and fire. My friends with automatics fumble and fight with thier camera because it is in some mode not apt to the photo at hand, with the camera refusing to activate the shutter. They miss the shot. I was able to just compose, focus and fire because it was a sunny day with little cloud cover, so I was able to take a light metering before hand, set my cameras shutter and apperture appropriately and then leave it for an hour or so without fear of over or underexposing! I can also meter seperately for direct sunlight and shade and then switch between the two with the apperture when needed. Very very effective techique that leaves hyper expensive automatics seem ineffectual and wasteful. Seeing someone run out of batteries is pretty funny too. My light meter battery last for a very long time and the batteries in my motor drive merely provide a convieniance.
People with "automatic" cameras continue to get erroneous results which would require them to correct or bracket (take 3 photos (for eg) with slightly differing metering and then at the end choose the best photo!), so what is the point of automatic! It is EASIER to do the lot than to figure out what the camera thinks of the scene and then manually adjusting.
Important point, understand the value of grey cards and hand held light meters! They are important and get around a huge problem with inbuilt camera meters. I'll try to explain it showing the extremes where error is worst...
When you have your camera pointing towards a scene, the scene may be composed of light shades (white, light greys, bright yellow, etc), or dark shades (black, dark greys, etc). The camera will assume (as they almost all do as part of their design), that the scene is composed of objects which on the whole reflect 18% of light (which appears as 50% grey). The end effect will be the bright scene and dark scene both coming out mid grey on average (in B&W). That means that the bright scene in underexposed and the dark scene overexposed.
This is because the camera cannot distiguish between bright objects dimmly lit and dark objects brightly lit. Here, both photos would also come out mid grey, when in reality, the bright objects should come out bright (high key scene) and the dark objects dark (low key scene) regardless of the amount of incident light present.
The RealTek 8139 PCI NIC redefines the meaning of 'low end.'
I've heard that the C and D variant fixed a lot of issues and I've had great success with them.
However, I'd like to nominate a contender. Davicom. I've had some extreme difficulties getting some Davicom NICs working at all without dropping connection or extreme speed problems. Under Win2k SP4, Debian Linux and OpenBSD. Admittedly these were all on-board in two classrooms I was challenged with. So the problems may have been with the implementation of the Davicom chipset and the motherboard.
They would fail to complete unicast, multicast or directed Ghosting sessions. Replacing with RTL8139D's gave me what I expected, solid 100Mbit ethernet.
Do you have anything to contribute? Because all you have shown is that you feel superior over a mix-up of simple English words, which I guess means you have a low self esteem due to feelings of general inferiority.
In addition, not everyone on the Internet has English as his or her first language. What matters is the intended message and not the spelling. Mixing "lose" and "loose" hardly renders the message unintelligible to intelligent people. So I guess you don't fit the description of intelligent then.
In the future, if x86 mobo's are DRM'ed to hell and back, just buy a decent computer with OpenFirmware.
Apple, Sun, etc.
After 20 years playing with computers, at home, the only "new" x86 machines are those I find thrown out on the streets. Apple and Sun gear is great and if Sun gives the 64bit AMD stuff OpenFirmware the picture will be complete.
This is getting out of hand, the last 30 or so comments have all been bsd is dying trolls. Maybe censorship isn't such a bad thing?
I didn't know about these particular trolls today until I read your post (I am browsing at 1). So I guess a person censoring what they want to see is probably best.
"develops software products to enhance the security, performance and user experience of Microsoft's Internet Information Services (IIS) Web server."
Entities who could be accused of having a conflict of interest, ought not bother at all with statements like these. It will only end up making them loose integrity.
They can tap your line all you want, but if they don't have the keys to your conversation, they're phucked.:-)
The encryption of GSM phones is only between the handset and the base station that you are currently connected to. The data is decrypted at the base station to be injected raw into what is typically digital packet switched PSTN nowdays.
The telco has the keys to your encryption with GSM, but does not need to go out of their way to actively use them to eavesdrop on you, because decryption is *required* to get onto the PSTN (it's a part of the system).
The psuedo random number generator used in GSM phones is a very weak sandwich of (3?) short linear feedback shift registers, which are initialized with your IMEI from memory (or something that simple).
This is designed to prevent any lay man with a radio scanner from listening to GSM phone calls. The fact that the transmission is digital, spread spectrum and encrypted makes eavesdropping difficult (but not impossible) to everyone but your telco/government.
Even if you call GSM to GSM and both handsets are connected to the same base station, still you are not assured of communications that are secure from your telco/gov or another very commited eavesdropper with enough money to break your weak security.
Re:Matrox put themselves in obscurity.
on
Video Card History
·
· Score: 1
Matrox does 2D and dual/multi monitor pretty well, but that's about it.
Definitely not a gaming card, though.
If you absolutely must have those extra few fps, then I guess Matrox are not the gaming cards for you. But this is silly. I played Half Life all the way through with a Matrox G200 and it was great.
This was back in the days when Tom's Hardware Guide was a whore for 3dfx and 3dfx did not offer an AGP card (later 3dfx offered a card with an AGP connector, which used the AGP slot as a PCI slot! Not using any AGP features due to their own chipset limitations.)
Bert McComas clearly shows either one of two things in his "technical article": 1. He does not have a technical grasp on what he writes and does not realise that failing to utilise one strong area of digital design (AGP) does not mean that that area is at fault. Or 2. He was intentionally mangling words to misrepresent the facts as if "AGP is worse than PCI" performance wise.
The fact is that AGP provided a bus that was many TIMES faster than PCI. Failing to utilise that bandwidth IS NOT THE FAULT OF AGP!
His comments fall down the toilet, where they belong, when Tom's hardware compared some AGP cards with PCI Voodoo2's, including a dual 12Mb Voodoo2 setup...
In a test with huge textures, textures too big to fully fit into the 12Mb local memory of the Voodoo2's, the Voodoo2's suddenly were at the complete mercy of the PCI bus.
The end result is a dual 12Mb Voodoo2 setup being completely beaten by some G200 (1.75 times faster) and TNT (2.6 times faster) cards!
AGP 1 PCI 0
Matrox cards have been more than usable for gaming.
OpenBSD 3.2, Pentium 75MHz, 32MB old 72pin EDO RAM, old narrow SCSI 520MB Seagate drive on some old Taiwanese VX motherboard I found thrown out during my local clean-up day:
Time to create 22MB file from/dev/zero: 24 seconds. Time to copy that file to same disk: 1 min 17 seconds. (plenty of old head thrashing).
firewall# dmesg OpenBSD 3.2 (GENERIC) #25: Thu Oct 3 19:51:53 MDT 2002
deraadt@i386.openbsd.org:/usr/src/sys/arch/i386/co mpile/GENERIC cpu0: F00F bug workaround installed cpu0: Intel Pentium (P54C) ("GenuineIntel" 586-class) 75 MHz cpu0: FPU,V86,DE,PSE,TSC,MSR,MCE,CX8 real mem = 33140736 (32364K) avail mem = 25227264 (24636K) using 430 buffers containing 1761280 bytes (1720K) of memory mainbus0 (root) bios0 at mainbus0: AT/286+(39) BIOS, date 07/10/96, BIOS32 rev. 0 @ 0xfb1e0 apm0 at bios0: Power Management spec V1.2 apm0: AC on, battery charge unknown, estimated 0:00 hours pcibios0 at bios0: rev. 2.1 @ 0xf0000/0xb700 pcibios0: PCI BIOS has 5 Interrupt Routing table entries pcibios0: PCI Interrupt Router at 000:07:0 ("Intel 82371SB PCI-ISA" rev 0x00) pcibios0: PCI bus #0 is the last bus bios0: ROM list: 0xdc000/0x4000 pci0 at mainbus0 bus 0: configuration mode 1 (bios) pchb0 at pci0 dev 0 function 0 "Intel 82437VX" rev 0x02 pcib0 at pci0 dev 7 function 0 "Intel 82371SB PCI-ISA" rev 0x01 pciide0 at pci0 dev 7 function 1 "Intel 82371SB IDE" rev 0x00: DMA, channel 0 wired to compatibility, channel 1 wired to compatibility pciide0: channel 0 ignored (disabled) pciide0: channel 1 ignored (disabled) rl0 at pci0 dev 20 function 0 "Realtek 8139" rev 0x10: irq 15 address 00:08:a1:28:72:e7 rlphy0 at rl0 phy 0: RTL internal phy isa0 at pcib0 isadma0 at isa0 pckbc0 at isa0 port 0x60/5 pckbd0 at pckbc0 (kbd slot) pckbc0: using irq 1 for kbd slot wskbd0 at pckbd0: console keyboard (bha probe): bha_cmd, cmd/data port empty 13 aha0 at isa0 port 0x330/4 irq 11 drq 5: model AHA-1542CF, firmware B.0 aha0: unlocking mailbox interface aha0: async, parity scsibus0 at aha0: 8 targets sd0 at scsibus0 targ 0 lun 0: [SEAGATE, ST3655N, 9550] SCSI2 0/direct fixed sd0: 520MB, 2493 cyl, 5 head, 85 sec, 512 bytes/sec, 1065036 sec total pcppi0 at isa0 port 0x61 midi0 at pcppi0: sysbeep0 at pcppi0 npx0 at isa0 port 0xf0/16: using exception 16 pccom0 at isa0 port 0x3f8/8 irq 4: ns16550a, 16 byte fifo pccom0: console pccom1 at isa0 port 0x2f8/8 irq 3: ns16550a, 16 byte fifo biomask 800 netmask 8800 ttymask 8802 pctr: 586-class performance counters and user-level cycle counter enabled dkcsum: sd0 matched BIOS disk 80 root on sd0a rootdev=0x400 rrootdev=0xd00 rawdev=0xd02 firewall# time dd bs=64k count=352 if=/dev/zero of=/22MB.bin 352+0 records in 352+0 records out 23068672 bytes transferred in 23.791 secs (969608 bytes/sec) 0.0u 2.0s 0:24.24 8.5% 0+0k 21+2490io 15pf+0w firewall# ls -la/22MB.bin -rw-r--r-- 1 root wheel 23068672 Oct 27 23:29/22MB.bin firewall# time cp/22MB.bin/home 0.0u 2.7s 1:16.37 3.5% 0+0k 379+2490io 11pf+0w firewall#
I don't want to start a holy war here, but what is the deal with you BSD fanatics? I've been sitting here at my freelance gig in front of a BSD box (a PIII 800 w/512 Megs of RAM) for about 20 minutes now while it attempts to copy a 17 Meg file from one folder on the hard drive to another folder. 20 minutes. At home, on my Pentium Pro 200 running NT 4, which by all standards should be a lot slower than this BSD box, the same operation would take about 2 minutes. If that.
From my old server...
login as: root root@10.0.0.2's password: Last login: Thu Oct 16 13:12:27 2003 from 10.0.0.13 OpenBSD 3.2 (GENERIC) #25: Thu Oct 3 19:51:53 MDT 2002
Welcome to OpenBSD: The proactively secure Unix-like operating system.
Please use the sendbug(1) utility to report bugs in the system. Before reporting a bug, please try to reproduce it with the latest version of the code. With bug reports, please try to ensure that enough information to reproduce the problem is enclosed, and if a known fix for it exists, include that as well.
Terminal type? [xterm] Don't login as root, use su oldserver# dmesg|grep cpu0 cpu0: F00F bug workaround installed cpu0: Intel Pentium/MMX ("GenuineIntel" 586-class) 200 MHz cpu0: FPU,V86,DE,PSE,TSC,MSR,MCE,CX8,MMX oldserver# dd if=/dev/zero of=/home/share/17MB.bin bs=32k count=544 544+0 records in 544+0 records out 17825792 bytes transferred in 1.110 secs (16057261 bytes/sec) oldserver# ls -la/home/share/*.bin -rw-r--r-- 1 root nobody 17825792 Oct 16 15:25/home/share/17MB.bin oldserver# time cp/home/share/17MB.bin/home/share/music 0.0u 1.3s 0:01.66 81.9% 0+0k 293+1916io 11pf+0w oldserver# time cp/home/share/17MB.bin/home/share/17MB.bak 0.0u 1.3s 0:05.23 26.0% 0+0k 286+1916io 0pf+0w
That's 1.66 seconds to copy a file exactly 17Mbytes from one disk to another, on an Pentium 200 MMX OpenBSD 3.2 machine.
To more closly match what you are doing, copying the same file to the same disk took 5.23 seconds.
1. You can not play games on it.Yeah, when I want to play games, the first thing I think is OpenBSD!
2. It cannot be used by my grandma.Neither can PIX. Your point?
3. It lacks a GUI of any note.You obviously consider any free Unix that doesn't come installed by default with a GUI as "lacking a GUI". I guess you use Mandrake because that is all you are capable of using.
5. It is an assortment of fragmented OSes.This is so untrue. The BSD's are whole and complete units in themselves. You wanna talk fragmentation, look at Linux.
8. Support for the latest hardware is always poor.BSD's often supports certain hardware before Linux does (crypto, USB, etc) and once something is supported, the support often tends to be much better (WiFi).
9. It is incompatiable with GNU/Linux.OpenBSD supports binary emulation of most programs from SVR4 (Solaris), FreeBSD, Linux, BSD/OS, SunOS and HP-UX. (from the OpenBSD front page)
10.It is dying.Yeah, right. Do you think that if you keep saying something, it will happen? The only thing that will happen, is that most people will ignore you and a few will take the time to ridicule you.
Maybe if someone with a maturity level higher than gradeschool takes over then OpenBSD has a good chance. But at the moment it's an unprofessional joke.
You base this opinion on your opinion of Theo?
I know of two major banks which use OpenBSD for their firewalls. I've received this info from insiders on independant occasions without asking (was a comment in passing, during conversation, regarding my use of OpenBSD).
The source is there and people do use OpenBSD's tree and their own private tree's for real, business critical work.
they'll think that one sounded better than the other.
Remember those green felt tipped pens you could buy at music shops? For drawing around the edge of the CD to prevent laser light leaving or re-enterin there or something? To "improve" the sound. ; )
Audiophile journalists rarely know what they're talking about. They do think that they have golden ears however.
How can something be "extended" and "natural" at the same time? Surely he means the the full range is available.
I remember once seeing a bunch of "respected" audiophile journalists (read elitist morons) on TV, in an A-B study of listening to CD's that have just come out of the freezer and the same CD's at room temperature.
Their claim, was that the CD manufacturing process causes cracks (I can't remember if these cracks were in the plastic or the aluminium) which caused distorted sound. Freezing the CD's, so they said, caused these cracks to come back together (while the CD remained cool) and thus the distortion disappeared.
They were trying to claim that CD does not sound as good as vinyl. Never mind the facts about vinyl that":
* Everything above 15kHz gets "rubbed off" after a couple of plays.
* Two channels are encoded into the one groove with an old analog process which then has to be decoded, compounding interchannel modulation, distortion and noise!
* Terrible wow and flutter, comparitively speaking.
I wish I could have been there to point out that if what they were saying were true, then the successful CDROM revolution has been a complete figment of the whole Worlds imagination.
I think it's pretty funny that there is this CD vs Vinyl war going on, when the reality is that the best analog tape decks are far better than the best vinyl units.
Large on-disk caches are not necessarily a good thing.
If you are using a decent OS which does a good job of caching disk activity, then on-disk cache is mostly a wasted delay (double caching).
What cache would you rather access? The dynamically variable cache in your systems main memory or the small cache at the end of your disk bus bottleneck?
On-disk cache can be good for read ahead data that has not been read yet, but I don't think the added delay in caching on-drive is worth it.
In fact, I've seen drive benchmarks at http://www.storagereview.com/ that show for example a base drive (with a smallish cache) slightly beating the special edition drive of the same type which just has a larger cache (usually 2Mb vs 8Mb).
And really, how often will you read the same 8Mb more than once before any other data retires it on a 300Gb drive?! And, if your OS didn't cache that, thus catching it and preventing the drive from having to do it at all, then you really have much bigger problems with your OS!
On drive caches, nowdays, are really an unecessary complexity.
Did anyone else here tirelessly transcribe HEX code from Compute! magazine into their C-64's to play games etc?
Compute! published some small BASIC code that was basically a checksuming program that would allow readers to input HEX code with a checksum at the end and beep if you made an error in that line.
I've heard that if Lithium is mixed with water, it explodes. Does anyone know if this is true?
I heard that an Australian army soldier was killed while walking in waist deep water when his (I guess typically badly treated) army radio battery got wet.
I'd love to know if this is true and Lithium is really *that* explosive?
WiFi uses non-ionizing radiation. After so many tests over so many years, scientists have very easily found ionizing radiation to cause or accelerate cancers, yet they have never been able to find evidence of the same with non-ionizing radiation.
Their kids are at infinitely more risk playing in the playground, soaking up UV from the Sun or sitting in front of a CRT which emits small ammounts of x rays.
What's more, typical WiFi puts out no more than 0.2W! Even if you have a highly directional antenna the effective radiated power is likely to be less than my Nokia 2110 (at 2 watts). Compare this with the absolute saturation of non-ionizing radiation that is all around us at much higher power levels from Cell phone, pager, TV, radio, etc etc!
As far as the G5 goes, what other PC, PC mind you, can you have 8-gigs of ram on or that comes stock wither SATA drives?! None yet. Apple is doing it right.
I've seen some Athlon XP motherboards which take up to 12GB of DDR.
Check into a program called "GetDataBack" - there are FAT and NTFS versions. It works very well.
I must support this statement. I've tried many different data recovery applications, including products which cost thousands and claim to be the market leaders.
Every single product I tried gave results ranging from useless to partial recovery. Except for GetDataBack, which has never failed to give me 100% recovery from filesystems which have even had partition tables and FAT's wiped out completely.
Absolute top tier and very very cheap too.
No consultancy should be without this product. You can more than recover the cost of the program with one client in need of data recovery.
Slashdot Mirror
I recently upgraded a friends PC from ME to XP Home. She purchased XP, which came with a sticker proclaiming that it included SP1a.
Since this was a recent purchase and the after thought SP1a sticker was there, I mistakenly assumed that it would be safe against Blaster.
Regardless, I enabled the built in firewall on the external interface NIC before I connected to the internet via her ADSL.
I couldn't get it going. I was using the ISP PPPoE driver which was supposed to work, but the ISP suggested I use the built in XP PPPoE driver, which worked fine. The phone tech also said that I must disable any firewall due to the use of a heartbeat initiated at their end.
So, I reluctantly did...
Her PC had Blaster literally within a minute or two of connecting.
But here comes the funny part... to get around the 60 seconds to shutdown, I double clicked the time to set the year back to give me a chance to remove the virus and patch her system. Unfortunately, during this, I had to reboot. At this stage the 30 day registration period was still in effect because I had not registered. Upon reboot, the 30 day period was up, XP was demanding I register now without giving me the desktop! Luckily it seems that it automatically connected.
Next time I'll just set it back an hour!
This kind of crap just has not happened to me on my Apple. In the end, I enabled the firewall and she has not had a problem. It might not have happened if I knew XP better (first install), but then I gave up on Microsoft long ago.
When they fall short, it is not a reflection of who I am, my intelligence or the size of my magic wand.
To sum up, it's not the wand, it's the wizard.
About 2 years ago I projected some images I took with some Agfa Scala 200 ASA film. Measured the size of the frame and the number of grains within a small square within that frame.
On average (the grains are random shapes and sizes, within limits), it came out to about 12 million grains within the full frame.
So it would take about 12 Mega pixels to match Scala 200 (maybe a little less than 12, to take into account the random nature of the grains which can make an image look less detailed due to the larger grains spoiling the effect of the whole). I imagine some of Fuji's highest resolution films would be 25 million grains or beyond.
I think the biggest hurdle to digital photography is not the CCD resolutions, but the available memory. Ever manipulated a high quality jpeg? The artifacts that were not noticable suddenly jump out at you when you run a filter, for example. So really high quality compression or lossless compression would be nice. A raw image at 25 Mpixels is a full 71 Megabytes for 24bit colour.
Fantastic advice.
Choose a make who builds great lenses. Nikon, Canon, Carl Zeis. Buy good lenses from that company (read reviews which do proper testing for resolution, distortion, contrast, etc, not just some subjective crap written by some self proclaimed knowledgable photographer) and then find a body to go with them. Buying expensive bodies and cheap lenses is a very bad decision. Spending $20,000 on lenses and $200 on a body is not a bad thing to do. The effects of the body with a good photographer are truely negligible as long as the shutter speeds are accurate.
If you really want to learn photography, get a fully manual camera with a built in light meter (to learn the flaws of built in light meters) and use either grey cards or a hand held light meter.
My personal favorite camera is my Nikon F2AS with the MD-2 motor drive.
Like the Pentax K-1000, it is built like a tank. Nikon takes two blocks of aluminium and machine the body and prism housing out of them. Very strong, 100% mechanical (besides the light meter). The AS variant is the best of the F2's which has a light meter, due to the sensor being silicon as opposed to the failure prone light dependant resistors of the previous models. The camera was released in 1977 and yet many current Nikkor lenses (and almost all Nikkor lenses) can be used with it.
I've been to places with friends where we've all tried to take a sudden "good shot", I've just had to raise my camera, focus and fire. My friends with automatics fumble and fight with thier camera because it is in some mode not apt to the photo at hand, with the camera refusing to activate the shutter. They miss the shot. I was able to just compose, focus and fire because it was a sunny day with little cloud cover, so I was able to take a light metering before hand, set my cameras shutter and apperture appropriately and then leave it for an hour or so without fear of over or underexposing! I can also meter seperately for direct sunlight and shade and then switch between the two with the apperture when needed. Very very effective techique that leaves hyper expensive automatics seem ineffectual and wasteful. Seeing someone run out of batteries is pretty funny too. My light meter battery last for a very long time and the batteries in my motor drive merely provide a convieniance.
People with "automatic" cameras continue to get erroneous results which would require them to correct or bracket (take 3 photos (for eg) with slightly differing metering and then at the end choose the best photo!), so what is the point of automatic! It is EASIER to do the lot than to figure out what the camera thinks of the scene and then manually adjusting.
Important point, understand the value of grey cards and hand held light meters! They are important and get around a huge problem with inbuilt camera meters. I'll try to explain it showing the extremes where error is worst...
When you have your camera pointing towards a scene, the scene may be composed of light shades (white, light greys, bright yellow, etc), or dark shades (black, dark greys, etc). The camera will assume (as they almost all do as part of their design), that the scene is composed of objects which on the whole reflect 18% of light (which appears as 50% grey). The end effect will be the bright scene and dark scene both coming out mid grey on average (in B&W). That means that the bright scene in underexposed and the dark scene overexposed.
This is because the camera cannot distiguish between bright objects dimmly lit and dark objects brightly lit. Here, both photos would also come out mid grey, when in reality, the bright objects should come out bright (high key scene) and the dark objects dark (low key scene) regardless of the amount of incident light present.
Metering for incident light with a hand
The RealTek 8139 PCI NIC redefines the meaning of 'low end.'
I've heard that the C and D variant fixed a lot of issues and I've had great success with them.
However, I'd like to nominate a contender. Davicom. I've had some extreme difficulties getting some Davicom NICs working at all without dropping connection or extreme speed problems. Under Win2k SP4, Debian Linux and OpenBSD. Admittedly these were all on-board in two classrooms I was challenged with. So the problems may have been with the implementation of the Davicom chipset and the motherboard.
They would fail to complete unicast, multicast or directed Ghosting sessions. Replacing with RTL8139D's gave me what I expected, solid 100Mbit ethernet.
How long until you idiots get it
Intelligence != knowledge.
It's called "English"
It's called a "Typo".
Do you have anything to contribute? Because all you have shown is that you feel superior over a mix-up of simple English words, which I guess means you have a low self esteem due to feelings of general inferiority.
In addition, not everyone on the Internet has English as his or her first language. What matters is the intended message and not the spelling. Mixing "lose" and "loose" hardly renders the message unintelligible to intelligent people. So I guess you don't fit the description of intelligent then.
Time to stock up on those old motherboards boys!
In the future, if x86 mobo's are DRM'ed to hell and back, just buy a decent computer with OpenFirmware.
Apple, Sun, etc.
After 20 years playing with computers, at home, the only "new" x86 machines are those I find thrown out on the streets. Apple and Sun gear is great and if Sun gives the 64bit AMD stuff OpenFirmware the picture will be complete.
censoring what they want to see
Oops, sorry. Should be "censoring what they don't want to see".
This is getting out of hand, the last 30 or so comments have all been bsd is dying trolls. Maybe censorship isn't such a bad thing?
I didn't know about these particular trolls today until I read your post (I am browsing at 1). So I guess a person censoring what they want to see is probably best.
"Netcraft is biased"
"develops software products to enhance the security, performance and user experience of Microsoft's Internet Information Services (IIS) Web server."
Entities who could be accused of having a conflict of interest, ought not bother at all with statements like these. It will only end up making them loose integrity.
They can tap your line all you want, but if they don't have the keys to your conversation, they're phucked. :-)
The encryption of GSM phones is only between the handset and the base station that you are currently connected to. The data is decrypted at the base station to be injected raw into what is typically digital packet switched PSTN nowdays.
The telco has the keys to your encryption with GSM, but does not need to go out of their way to actively use them to eavesdrop on you, because decryption is *required* to get onto the PSTN (it's a part of the system).
The psuedo random number generator used in GSM phones is a very weak sandwich of (3?) short linear feedback shift registers, which are initialized with your IMEI from memory (or something that simple).
This is designed to prevent any lay man with a radio scanner from listening to GSM phone calls. The fact that the transmission is digital, spread spectrum and encrypted makes eavesdropping difficult (but not impossible) to everyone but your telco/government.
Even if you call GSM to GSM and both handsets are connected to the same base station, still you are not assured of communications that are secure from your telco/gov or another very commited eavesdropper with enough money to break your weak security.
Matrox does 2D and dual/multi monitor pretty well, but that's about it.
Definitely not a gaming card, though.
If you absolutely must have those extra few fps, then I guess Matrox are not the gaming cards for you. But this is silly. I played Half Life all the way through with a Matrox G200 and it was great.
Tom and Bert McComas would have you beleive that AGP gives WORSE performance than PCI!
This was back in the days when Tom's Hardware Guide was a whore for 3dfx and 3dfx did not offer an AGP card (later 3dfx offered a card with an AGP connector, which used the AGP slot as a PCI slot! Not using any AGP features due to their own chipset limitations.)
Bert McComas clearly shows either one of two things in his "technical article": 1. He does not have a technical grasp on what he writes and does not realise that failing to utilise one strong area of digital design (AGP) does not mean that that area is at fault. Or 2. He was intentionally mangling words to misrepresent the facts as if "AGP is worse than PCI" performance wise.
The fact is that AGP provided a bus that was many TIMES faster than PCI. Failing to utilise that bandwidth IS NOT THE FAULT OF AGP!
His comments fall down the toilet, where they belong, when Tom's hardware compared some AGP cards with PCI Voodoo2's, including a dual 12Mb Voodoo2 setup...
In a test with huge textures, textures too big to fully fit into the 12Mb local memory of the Voodoo2's, the Voodoo2's suddenly were at the complete mercy of the PCI bus.
The end result is a dual 12Mb Voodoo2 setup being completely beaten by some G200 (1.75 times faster) and TNT (2.6 times faster) cards!
AGP 1
PCI 0
Matrox cards have been more than usable for gaming.
Diamond was hot for awhile
I remember when the Diamond Speedstar 24X was the fastest 2D card around.
It all seems so ridiculous now. : )
I wonder if the Australian police accidentally stormed the next door neighbours and shot the innocent occupants in the face with a 12 gauge shotgun?
As has happened in the past, by the now disbanded Tactical Response Group.
To add to the anti Trolling...
/dev/zero: 24 seconds.
o mpile/GENERIC /22MB.bin /22MB.bin /22MB.bin /home
OpenBSD 3.2, Pentium 75MHz, 32MB old 72pin EDO RAM, old narrow SCSI 520MB Seagate drive on some old Taiwanese VX motherboard I found thrown out during my local clean-up day:
Time to create 22MB file from
Time to copy that file to same disk: 1 min 17 seconds. (plenty of old head thrashing).
firewall# dmesg
OpenBSD 3.2 (GENERIC) #25: Thu Oct 3 19:51:53 MDT 2002
deraadt@i386.openbsd.org:/usr/src/sys/arch/i386/c
cpu0: F00F bug workaround installed
cpu0: Intel Pentium (P54C) ("GenuineIntel" 586-class) 75 MHz
cpu0: FPU,V86,DE,PSE,TSC,MSR,MCE,CX8
real mem = 33140736 (32364K)
avail mem = 25227264 (24636K)
using 430 buffers containing 1761280 bytes (1720K) of memory
mainbus0 (root)
bios0 at mainbus0: AT/286+(39) BIOS, date 07/10/96, BIOS32 rev. 0 @ 0xfb1e0
apm0 at bios0: Power Management spec V1.2
apm0: AC on, battery charge unknown, estimated 0:00 hours
pcibios0 at bios0: rev. 2.1 @ 0xf0000/0xb700
pcibios0: PCI BIOS has 5 Interrupt Routing table entries
pcibios0: PCI Interrupt Router at 000:07:0 ("Intel 82371SB PCI-ISA" rev 0x00)
pcibios0: PCI bus #0 is the last bus
bios0: ROM list: 0xdc000/0x4000
pci0 at mainbus0 bus 0: configuration mode 1 (bios)
pchb0 at pci0 dev 0 function 0 "Intel 82437VX" rev 0x02
pcib0 at pci0 dev 7 function 0 "Intel 82371SB PCI-ISA" rev 0x01
pciide0 at pci0 dev 7 function 1 "Intel 82371SB IDE" rev 0x00: DMA, channel 0 wired to compatibility, channel 1 wired to compatibility
pciide0: channel 0 ignored (disabled)
pciide0: channel 1 ignored (disabled)
rl0 at pci0 dev 20 function 0 "Realtek 8139" rev 0x10: irq 15 address 00:08:a1:28:72:e7
rlphy0 at rl0 phy 0: RTL internal phy
isa0 at pcib0
isadma0 at isa0
pckbc0 at isa0 port 0x60/5
pckbd0 at pckbc0 (kbd slot)
pckbc0: using irq 1 for kbd slot
wskbd0 at pckbd0: console keyboard
(bha probe): bha_cmd, cmd/data port empty 13
aha0 at isa0 port 0x330/4 irq 11 drq 5: model AHA-1542CF, firmware B.0
aha0: unlocking mailbox interface
aha0: async, parity
scsibus0 at aha0: 8 targets
sd0 at scsibus0 targ 0 lun 0: [SEAGATE, ST3655N, 9550] SCSI2 0/direct fixed
sd0: 520MB, 2493 cyl, 5 head, 85 sec, 512 bytes/sec, 1065036 sec total
pcppi0 at isa0 port 0x61
midi0 at pcppi0:
sysbeep0 at pcppi0
npx0 at isa0 port 0xf0/16: using exception 16
pccom0 at isa0 port 0x3f8/8 irq 4: ns16550a, 16 byte fifo
pccom0: console
pccom1 at isa0 port 0x2f8/8 irq 3: ns16550a, 16 byte fifo
biomask 800 netmask 8800 ttymask 8802
pctr: 586-class performance counters and user-level cycle counter enabled
dkcsum: sd0 matched BIOS disk 80
root on sd0a
rootdev=0x400 rrootdev=0xd00 rawdev=0xd02
firewall# time dd bs=64k count=352 if=/dev/zero of=/22MB.bin
352+0 records in
352+0 records out
23068672 bytes transferred in 23.791 secs (969608 bytes/sec)
0.0u 2.0s 0:24.24 8.5% 0+0k 21+2490io 15pf+0w
firewall# ls -la
-rw-r--r-- 1 root wheel 23068672 Oct 27 23:29
firewall# time cp
0.0u 2.7s 1:16.37 3.5% 0+0k 379+2490io 11pf+0w
firewall#
I don't want to start a holy war here, but what is the deal with you BSD fanatics? I've been sitting here at my freelance gig in front of a BSD box (a PIII 800 w/512 Megs of RAM) for about 20 minutes now while it attempts to copy a 17 Meg file from one folder on the hard drive to another folder. 20 minutes. At home, on my Pentium Pro 200 running NT 4, which by all standards should be a lot slower than this BSD box, the same operation would take about 2 minutes. If that.
/home/share/*.bin /home/share/17MB.bin /home/share/17MB.bin /home/share/music /home/share/17MB.bin /home/share/17MB.bak
a dd&sektion=1&format=html
From my old server...
login as: root
root@10.0.0.2's password:
Last login: Thu Oct 16 13:12:27 2003 from 10.0.0.13
OpenBSD 3.2 (GENERIC) #25: Thu Oct 3 19:51:53 MDT 2002
Welcome to OpenBSD: The proactively secure Unix-like operating system.
Please use the sendbug(1) utility to report bugs in the system.
Before reporting a bug, please try to reproduce it with the latest
version of the code. With bug reports, please try to ensure that
enough information to reproduce the problem is enclosed, and if a
known fix for it exists, include that as well.
Terminal type? [xterm]
Don't login as root, use su
oldserver# dmesg|grep cpu0
cpu0: F00F bug workaround installed
cpu0: Intel Pentium/MMX ("GenuineIntel" 586-class) 200 MHz
cpu0: FPU,V86,DE,PSE,TSC,MSR,MCE,CX8,MMX
oldserver# dd if=/dev/zero of=/home/share/17MB.bin bs=32k count=544
544+0 records in
544+0 records out
17825792 bytes transferred in 1.110 secs (16057261 bytes/sec)
oldserver# ls -la
-rw-r--r-- 1 root nobody 17825792 Oct 16 15:25
oldserver# time cp
0.0u 1.3s 0:01.66 81.9% 0+0k 293+1916io 11pf+0w
oldserver# time cp
0.0u 1.3s 0:05.23 26.0% 0+0k 286+1916io 0pf+0w
That's 1.66 seconds to copy a file exactly 17Mbytes from one disk to another, on an Pentium 200 MMX OpenBSD 3.2 machine.
To more closly match what you are doing, copying the same file to the same disk took 5.23 seconds.
1. You can not play games on it. Yeah, when I want to play games, the first thing I think is OpenBSD!
2. It cannot be used by my grandma. Neither can PIX. Your point?
3. It lacks a GUI of any note. You obviously consider any free Unix that doesn't come installed by default with a GUI as "lacking a GUI". I guess you use Mandrake because that is all you are capable of using.
4. There is no support available for it. http://www.openbsd.org/support.html
5. It is an assortment of fragmented OSes. This is so untrue. The BSD's are whole and complete units in themselves. You wanna talk fragmentation, look at Linux.
6. It cannot be run on the x86 platform. Idiot. This is the worst troll I have ever seen. http://www.openbsd.org/plat.html
7. You have to compile everything and know C. http://www.openbsd.org/cgi-bin/man.cgi?query=pkg_
8. Support for the latest hardware is always poor. BSD's often supports certain hardware before Linux does (crypto, USB, etc) and once something is supported, the support often tends to be much better (WiFi).
9. It is incompatiable with GNU/Linux. OpenBSD supports binary emulation of most programs from SVR4 (Solaris), FreeBSD, Linux, BSD/OS, SunOS and HP-UX. (from the OpenBSD front page)
10.It is dying. Yeah, right. Do you think that if you keep saying something, it will happen? The only thing that will happen, is that most people will ignore you and a few will take the time to ridicule you.
Maybe if someone with a maturity level higher than gradeschool takes over then OpenBSD has a good chance. But at the moment it's an unprofessional joke.
You base this opinion on your opinion of Theo?
I know of two major banks which use OpenBSD for their firewalls. I've received this info from insiders on independant occasions without asking (was a comment in passing, during conversation, regarding my use of OpenBSD).
The source is there and people do use OpenBSD's tree and their own private tree's for real, business critical work.
they'll think that one sounded better than the other.
Remember those green felt tipped pens you could buy at music shops? For drawing around the edge of the CD to prevent laser light leaving or re-enterin there or something? To "improve" the sound. ; )
Audiophile journalists rarely know what they're talking about. They do think that they have golden ears however.
How can something be "extended" and "natural" at the same time? Surely he means the the full range is available.
I remember once seeing a bunch of "respected" audiophile journalists (read elitist morons) on TV, in an A-B study of listening to CD's that have just come out of the freezer and the same CD's at room temperature.
Their claim, was that the CD manufacturing process causes cracks (I can't remember if these cracks were in the plastic or the aluminium) which caused distorted sound. Freezing the CD's, so they said, caused these cracks to come back together (while the CD remained cool) and thus the distortion disappeared.
They were trying to claim that CD does not sound as good as vinyl. Never mind the facts about vinyl that":
* Everything above 15kHz gets "rubbed off" after a couple of plays.
* Two channels are encoded into the one groove with an old analog process which then has to be decoded, compounding interchannel modulation, distortion and noise!
* Terrible wow and flutter, comparitively speaking.
I wish I could have been there to point out that if what they were saying were true, then the successful CDROM revolution has been a complete figment of the whole Worlds imagination.
I think it's pretty funny that there is this CD vs Vinyl war going on, when the reality is that the best analog tape decks are far better than the best vinyl units.
Large on-disk caches are not necessarily a good thing.
If you are using a decent OS which does a good job of caching disk activity, then on-disk cache is mostly a wasted delay (double caching).
What cache would you rather access? The dynamically variable cache in your systems main memory or the small cache at the end of your disk bus bottleneck?
On-disk cache can be good for read ahead data that has not been read yet, but I don't think the added delay in caching on-drive is worth it.
In fact, I've seen drive benchmarks at http://www.storagereview.com/ that show for example a base drive (with a smallish cache) slightly beating the special edition drive of the same type which just has a larger cache (usually 2Mb vs 8Mb).
And really, how often will you read the same 8Mb more than once before any other data retires it on a 300Gb drive?! And, if your OS didn't cache that, thus catching it and preventing the drive from having to do it at all, then you really have much bigger problems with your OS!
On drive caches, nowdays, are really an unecessary complexity.
Did anyone else here tirelessly transcribe HEX code from Compute! magazine into their C-64's to play games etc?
Compute! published some small BASIC code that was basically a checksuming program that would allow readers to input HEX code with a checksum at the end and beep if you made an error in that line.
Pages and pages and pages of HEX! Arghhhhh!
I've heard that if Lithium is mixed with water, it explodes. Does anyone know if this is true?
I heard that an Australian army soldier was killed while walking in waist deep water when his (I guess typically badly treated) army radio battery got wet.
I'd love to know if this is true and Lithium is really *that* explosive?
and set a completely unfair precedent.
WiFi uses non-ionizing radiation. After so many tests over so many years, scientists have very easily found ionizing radiation to cause or accelerate cancers, yet they have never been able to find evidence of the same with non-ionizing radiation.
Their kids are at infinitely more risk playing in the playground, soaking up UV from the Sun or sitting in front of a CRT which emits small ammounts of x rays.
What's more, typical WiFi puts out no more than 0.2W! Even if you have a highly directional antenna the effective radiated power is likely to be less than my Nokia 2110 (at 2 watts). Compare this with the absolute saturation of non-ionizing radiation that is all around us at much higher power levels from Cell phone, pager, TV, radio, etc etc!
This is a crazy lawsuit.
As far as the G5 goes, what other PC, PC mind you, can you have 8-gigs of ram on or that comes stock wither SATA drives?! None yet. Apple is doing it right.
I've seen some Athlon XP motherboards which take up to 12GB of DDR.
Check into a program called "GetDataBack" - there are FAT and NTFS versions. It works very well.
I must support this statement. I've tried many different data recovery applications, including products which cost thousands and claim to be the market leaders.
Every single product I tried gave results ranging from useless to partial recovery. Except for GetDataBack, which has never failed to give me 100% recovery from filesystems which have even had partition tables and FAT's wiped out completely.
Absolute top tier and very very cheap too.
No consultancy should be without this product. You can more than recover the cost of the program with one client in need of data recovery.