I agree, I think we do need Apple, or at least we did.
They kickstarted the horribly lagging USB device market. A lot of windows GUI elements seem to come from Apple. Zen seems to be heavily iPod inspired, for all I know, maybe we'd be stuck with ugly Nomads and Nomad clones.
Before Jobs took over, the PC market looked like a bunch of ugly square beige boxes, since, we now get a bunch of ugly multi-colored varying swoop-shaped windowed boxes. I haven't checked to see if there are any G5 inspired PC cases yet.
f you feel you have been wronged by because your 1,000,000 hour MTBF drive will only last 900,000 hours
Huh? It's far worse than that! More like a failure within 1000 to 5000 hours for many people. I knew a guy that maintained a 18 node cluster that had to send in half of its IBM drives, and all the replacements failed too!
Buran was expensive crap (like US shuttle, just a little cheaper) but Energia is a good lifter. Buran and the US Shuttle are too big and too expensive for the people moving task.
The US shuttle certainly was more glamorous than the soviet method of keeping people movier / cargo hauler separate. The only advantage to a shuttle method is to repair, upgrade and even bring back satellites.
Unfortunately, law and technology are completely different realms. The law realm tries its damndest to be exacting which costs expedience, and the technology realm tries its damndest to be fast.
Gates said, no one will need more than 640k of memory
This is apocryphal at best. It is really an urban legend that he said that.
And I think you are misunderstanding the grandparent's post. Very few people need 64bit now. Eventually, yes, there will be a changeover. IIRC, Intel was projecting 2005/2006. No "desktop" that I know of now needs more than 4GB RAM, few workstations need that much as well.
The main benefit of AMD's 64 bit implementation is that it adds registers to the register starved x86 architecture, and this is enough to more than offset the extra bandwidth and memory wasted on 64 bit pointers. The ability to do cryptography quicker generally wasn't enough to offset the penalties of a longer pointer, unless you happen to do cryptography as your livelyhood.
I do like other things about the AMD chip though, like the cheaper point-to-point interconnect. The EV6 bus took too many wires to do cheap multiprocessing, and of course, Intel is using the multi-drop bus, which effectively limited the bandwidth available to Xeon and Itanium.
The problem is that the digital distribution shifts the cost of distribution away from the studios and to the theater owners. Per copy distribution is much lower but it requires recieving and storage hardware and of course, getting $100K+ digital projectors in an attempt to replace still functioning and paid-for film projectors.
I'm betting that the studios are trying to pocket the entire difference rather than allowing the theater owners a break for having to invest in expensive hardware.
The down side of the desktop is that it pretty much stays on the top of your desk
The other downside is that they are big. If you get an LCD panel then you are much closer to the laptop pricing anyway. Dorms don't have much room, and the ability to put the computer away when not needed is important to some.
The business case for tablets assumes that the end user either a) doesn't know how to type, or b) isn't in a position to type.
I think replacing typing is a sideline, and I'd get the keyboard with it anyway. IIRC, the business case is about easily adding sketches. The ads I see don't really promote handwriting recognition much compared to the amount they promote the ability to easily mix drawings and text. This sort of thing would be handy for lecture notes in just about any study discipline other than maybe strict literature.
I can tell you will be able to plug a PEG gfx card into ANY PEG slot on your board
Depends on what the other slots are.
The prototype boards for BTX format show only one x16 slot in the system, and the rest of the slots were x1, plus I think one "legacy" PCI card. If the card is designed for x16, you need an x16 slot to plug into it.
So? Prices almost always come down regardless of a change in standards.
And I think AGP will still be supported for a while. Evidence? It's still not hard to find PCI video cards half a decade after the standard was supersceeded. They don't stack up against AGP 4x/8x but they are still readily available and I think are still manufactured. The only wrinkle is that PCI is still available on motherboards, IIRC, eventually neither AGP nor PCI will be on PCI-Express boards.
I really don't think the value of old AGP motherboards will rise, I really don't see the point in clinging to an old standard that hard. I don't see the value of ISA/EISA mainboards rising, except for the fact that the cheaper retailers sell out of their stock, and the pricier ones may still have new-old-stock ones just because people refuse to pay the price.
Uh, I thought most video cards now are multi-head.
I knew that dual opteron boards were expensive, but I thought the price was in PCI-X. One can get dual opteron with PCI & AGP for maybe less than half with PCI-X. The only dual opteron boards I see around $200 only have PCI-33/32 that I can tell.
Tyan Thunder K8W runs $450, and that has four memory channels, two PCI-X busses and AGP. Tyan Thunder K8S, very similar board but without AGP, runs $520. Both prices are at Newegg.
I actually think the prices are very fair considering what's packed into the boards, that is probably 2x to 3x the number of wires vs. the cheapie consumer boards, I mean, you get what you pay for. The cheapest dual opteron with PCI-X with or without AGP is $380.
A side note (not necessarily for parent poster): For those that haven't noticed, PCI-X is different from PCI-Express. PCI-X is an extension of PCI, PCI-Express is a totally different creature.
Well, I assumed that AGP was supposed allow bi-directional data transfer. I suppose there are caveats somewhere.
buying a top of the line AGP card at this juncture is equally ridiculous...
Buying top-of-the-line is generally rediculous, period, unless you really do need it. It is best to wait six months after initial release for prices to come down esp. when a replacement comes out, and also for more polished drivers.
It really doesn't affect me, I'll buy what I need and what works for my current system because there's plenty of expansion potential left. There are some benefits to PCI-E but I have PCI 66/64 available which suits just fine for now.
Re:AMD have been better than Intel for some time..
on
AMD Back in the Black
·
· Score: 1
I've stuck with Intel because of the IMO Athlon's poorer heatsink mounting mechanism and the exposed core. I've known people that cracked a core trying to mount the heatsink.
IMO, it wasn't until the A64 line when AMD really could compete well against Intel on performance, before then it was mostly just price.
Also, too many of the Athlon chipsets had IMO poor PCI implementations, particularly at busmastering. AMD's own chipsets were better at this, I have a few pieces of hardware that required workarounds for Via and SIS compatibility. Even if the performance lagged, I'd prefer Intel's or AMD's chipsets rather than other chips that might be incompatible with boards I might buy in the future.
Buffy had a good seven year run. According to the commentaries, Joss had originally planned to end it at the end of the fifth, although I didn't get to the part on why he wanted it to end there.
Now, I'm only half way through Angel Season 1.
I've never bothered with the broadcast because of reception and ad issues.
I'm sorry, the two Sager laptops I've personally seen are shit. They are kind of flimsy, look plasticky and all around, IMO, not durable. Even the Dells I've seen were twice as good construction quality, IMO.
It's bad enough that this sort of thing is sort of thing is exploited for PR. Remember the company associated with the Raeleans? The homely PR woman tried to reassure that outside scientiests will be allowed to verify their claims. Where are they now? Where is the verification? What happened to the clone?
Interesting page. I'd say that your guess is probably right. I think Jobs did promise 3GHz G5s sometime mid to late 2004. For all we know, Big Blue is stockpiling some 2.5GHz rated chips right now.
I'd sort of tend to agree, although under your standards, the stuff I do as an EE really would fit under development, we don't have the budget to send out for external certification and external testing. No biggie, I guess I can live with being a hardware developer.
Is it true that some states have prohibited Microsoft from issuing MSCEs? I heard this somewhere but I can't remember. Something about Microsoft not having the authority to certify engineers.
I think that's stretching it a lot because you don't show a convincing causality. It doesn't show that the decision to go L2 was made because it avoids costs of upgrades vs. technical reasons which happen to avoid any chance of servicing.
The infrared noise issue is convincing enough of a reason because you want everything to be not just cold but damn cold to maintain the lowest noise floor. The telescope will not be able to measure temperatures that are below that of the measuring instrument's sensor temperature.
Honestly though, someone needs to do something about lamp life. They've made some pretty good strides with decreasing bulb cost, but I still don't want to be replacing a bulb every few months.
You can run a 2000 hour bulb 40 hours a week and it will still last 50 weeks.
With my usage, I think my projector's bulb will last another two years.
There are some front projectors that have 3000 hr bulbs, a few even have 5000 hour ratings. Sometimes one can trade brightness for bulb life.
What I will do when the bulb runs out, I don't know because so much can change. It's possible that some dark horse can arrive on the scene, heck, even Intel is thinking of getting into LCOS chip fabbing. While not as techically sexy as DLP's micromirrors, I think there's some room for improvement in those devices, so one can get a decent picture without DLP's rainbows.
I suppose it's possible that with TI's DLP patents running out, maybe we'll even see three-chip DLP units for home theater and that would squash my biggest complaint about it, the rainbows.
Has the "uncanny valley" been proven? My understanding is that the entire theory is conjecture at best without an experimental basis because there wasn't any good enough equipment to perform such an experiment.
I do find it highly odd that some people have used this theory as their reason to avoid this research, at least one good point of this research is to help prove or disprove it. Fine with me. It doesn't matter to me what most experts believe because experts can be highly qualified, reputable and still be on the wrong "side".
It looks like even the mobile workstations (expensive laptops) seem to require Windows licence. Last I checked, Dell and HP sold desktop workstations without Linux, but the catch is that it often costs an extra $100 vs. the Windows version.
Slashdot Mirror
I agree, I think we do need Apple, or at least we did.
They kickstarted the horribly lagging USB device market. A lot of windows GUI elements seem to come from Apple. Zen seems to be heavily iPod inspired, for all I know, maybe we'd be stuck with ugly Nomads and Nomad clones.
Before Jobs took over, the PC market looked like a bunch of ugly square beige boxes, since, we now get a bunch of ugly multi-colored varying swoop-shaped windowed boxes. I haven't checked to see if there are any G5 inspired PC cases yet.
f you feel you have been wronged by because your 1,000,000 hour MTBF drive will only last 900,000 hours
Huh? It's far worse than that! More like a failure within 1000 to 5000 hours for many people. I knew a guy that maintained a 18 node cluster that had to send in half of its IBM drives, and all the replacements failed too!
Buran was expensive crap (like US shuttle, just a little cheaper) but Energia is a good lifter. Buran and the US Shuttle are too big and too expensive for the people moving task.
The US shuttle certainly was more glamorous than the soviet method of keeping people movier / cargo hauler separate. The only advantage to a shuttle method is to repair, upgrade and even bring back satellites.
Unfortunately, law and technology are completely different realms. The law realm tries its damndest to be exacting which costs expedience, and the technology realm tries its damndest to be fast.
Gates said, no one will need more than 640k of memory
This is apocryphal at best. It is really an urban legend that he said that.
And I think you are misunderstanding the grandparent's post. Very few people need 64bit now. Eventually, yes, there will be a changeover. IIRC, Intel was projecting 2005/2006. No "desktop" that I know of now needs more than 4GB RAM, few workstations need that much as well.
The main benefit of AMD's 64 bit implementation is that it adds registers to the register starved x86 architecture, and this is enough to more than offset the extra bandwidth and memory wasted on 64 bit pointers. The ability to do cryptography quicker generally wasn't enough to offset the penalties of a longer pointer, unless you happen to do cryptography as your livelyhood.
I do like other things about the AMD chip though, like the cheaper point-to-point interconnect. The EV6 bus took too many wires to do cheap multiprocessing, and of course, Intel is using the multi-drop bus, which effectively limited the bandwidth available to Xeon and Itanium.
Sweet Don Quixote, why can't they get it through their heads that higher quality is more important than higher profit?
Heh, this is America we're talking about.
American companies gladly halving reliability and/or doubling danger to save five bucks on a car, while paying their CEOs into the nine digits.
Hmm...ultimately the customers will.
In a way, yes.
The problem is that the digital distribution shifts the cost of distribution away from the studios and to the theater owners. Per copy distribution is much lower but it requires recieving and storage hardware and of course, getting $100K+ digital projectors in an attempt to replace still functioning and paid-for film projectors.
I'm betting that the studios are trying to pocket the entire difference rather than allowing the theater owners a break for having to invest in expensive hardware.
The down side of the desktop is that it pretty much stays on the top of your desk
The other downside is that they are big. If you get an LCD panel then you are much closer to the laptop pricing anyway. Dorms don't have much room, and the ability to put the computer away when not needed is important to some.
The business case for tablets assumes that the end user either a) doesn't know how to type, or b) isn't in a position to type.
I think replacing typing is a sideline, and I'd get the keyboard with it anyway. IIRC, the business case is about easily adding sketches. The ads I see don't really promote handwriting recognition much compared to the amount they promote the ability to easily mix drawings and text. This sort of thing would be handy for lecture notes in just about any study discipline other than maybe strict literature.
I can tell you will be able to plug a PEG gfx card into ANY PEG slot on your board
Depends on what the other slots are.
The prototype boards for BTX format show only one x16 slot in the system, and the rest of the slots were x1, plus I think one "legacy" PCI card. If the card is designed for x16, you need an x16 slot to plug into it.
So? Prices almost always come down regardless of a change in standards.
And I think AGP will still be supported for a while. Evidence? It's still not hard to find PCI video cards half a decade after the standard was supersceeded. They don't stack up against AGP 4x/8x but they are still readily available and I think are still manufactured. The only wrinkle is that PCI is still available on motherboards, IIRC, eventually neither AGP nor PCI will be on PCI-Express boards.
I really don't think the value of old AGP motherboards will rise, I really don't see the point in clinging to an old standard that hard. I don't see the value of ISA/EISA mainboards rising, except for the fact that the cheaper retailers sell out of their stock, and the pricier ones may still have new-old-stock ones just because people refuse to pay the price.
Uh, I thought most video cards now are multi-head.
I knew that dual opteron boards were expensive, but I thought the price was in PCI-X. One can get dual opteron with PCI & AGP for maybe less than half with PCI-X. The only dual opteron boards I see around $200 only have PCI-33/32 that I can tell.
Tyan Thunder K8W runs $450, and that has four memory channels, two PCI-X busses and AGP. Tyan Thunder K8S, very similar board but without AGP, runs $520. Both prices are at Newegg.
I actually think the prices are very fair considering what's packed into the boards, that is probably 2x to 3x the number of wires vs. the cheapie consumer boards, I mean, you get what you pay for. The cheapest dual opteron with PCI-X with or without AGP is $380.
A side note (not necessarily for parent poster): For those that haven't noticed, PCI-X is different from PCI-Express. PCI-X is an extension of PCI, PCI-Express is a totally different creature.
Well, I assumed that AGP was supposed allow bi-directional data transfer. I suppose there are caveats somewhere.
buying a top of the line AGP card at this juncture is equally ridiculous...
Buying top-of-the-line is generally rediculous, period, unless you really do need it. It is best to wait six months after initial release for prices to come down esp. when a replacement comes out, and also for more polished drivers.
It really doesn't affect me, I'll buy what I need and what works for my current system because there's plenty of expansion potential left. There are some benefits to PCI-E but I have PCI 66/64 available which suits just fine for now.
I've stuck with Intel because of the IMO Athlon's poorer heatsink mounting mechanism and the exposed core. I've known people that cracked a core trying to mount the heatsink.
IMO, it wasn't until the A64 line when AMD really could compete well against Intel on performance, before then it was mostly just price.
Also, too many of the Athlon chipsets had IMO poor PCI implementations, particularly at busmastering. AMD's own chipsets were better at this, I have a few pieces of hardware that required workarounds for Via and SIS compatibility. Even if the performance lagged, I'd prefer Intel's or AMD's chipsets rather than other chips that might be incompatible with boards I might buy in the future.
Buffy had a good seven year run. According to the commentaries, Joss had originally planned to end it at the end of the fifth, although I didn't get to the part on why he wanted it to end there.
Now, I'm only half way through Angel Season 1.
I've never bothered with the broadcast because of reception and ad issues.
I'm sorry, the two Sager laptops I've personally seen are shit. They are kind of flimsy, look plasticky and all around, IMO, not durable. Even the Dells I've seen were twice as good construction quality, IMO.
It's bad enough that this sort of thing is sort of thing is exploited for PR. Remember the company associated with the Raeleans? The homely PR woman tried to reassure that outside scientiests will be allowed to verify their claims. Where are they now? Where is the verification? What happened to the clone?
Hmm, correction in italics:
For all we know, Big Blue is probably stockpiling some 2.5GHz rated chips right now.
Interesting page. I'd say that your guess is probably right. I think Jobs did promise 3GHz G5s sometime mid to late 2004. For all we know, Big Blue is stockpiling some 2.5GHz rated chips right now.
I'd sort of tend to agree, although under your standards, the stuff I do as an EE really would fit under development, we don't have the budget to send out for external certification and external testing. No biggie, I guess I can live with being a hardware developer.
Is it true that some states have prohibited Microsoft from issuing MSCEs? I heard this somewhere but I can't remember. Something about Microsoft not having the authority to certify engineers.
I knew about the rovers but not much of a deal was made about them in many of the magazine articles that mentioned them, or many of the books I read.
I guess actually sending people to drive/ride on a huge rover was more interesting.
There's your economic reason.
I think that's stretching it a lot because you don't show a convincing causality. It doesn't show that the decision to go L2 was made because it avoids costs of upgrades vs. technical reasons which happen to avoid any chance of servicing.
The infrared noise issue is convincing enough of a reason because you want everything to be not just cold but damn cold to maintain the lowest noise floor. The telescope will not be able to measure temperatures that are below that of the measuring instrument's sensor temperature.
Apart from huge savings, it allows US companies to concentrate on their core competencies
Uh oh. US companies don't know what their core competencies are.
Comment adapted from a Dilbert comic strip.
Honestly though, someone needs to do something about lamp life. They've made some pretty good strides with decreasing bulb cost, but I still don't want to be replacing a bulb every few months.
You can run a 2000 hour bulb 40 hours a week and it will still last 50 weeks.
With my usage, I think my projector's bulb will last another two years.
There are some front projectors that have 3000 hr bulbs, a few even have 5000 hour ratings. Sometimes one can trade brightness for bulb life.
What I will do when the bulb runs out, I don't know because so much can change. It's possible that some dark horse can arrive on the scene, heck, even Intel is thinking of getting into LCOS chip fabbing. While not as techically sexy as DLP's micromirrors, I think there's some room for improvement in those devices, so one can get a decent picture without DLP's rainbows.
I suppose it's possible that with TI's DLP patents running out, maybe we'll even see three-chip DLP units for home theater and that would squash my biggest complaint about it, the rainbows.
Has the "uncanny valley" been proven? My understanding is that the entire theory is conjecture at best without an experimental basis because there wasn't any good enough equipment to perform such an experiment.
I do find it highly odd that some people have used this theory as their reason to avoid this research, at least one good point of this research is to help prove or disprove it. Fine with me. It doesn't matter to me what most experts believe because experts can be highly qualified, reputable and still be on the wrong "side".
It looks like even the mobile workstations (expensive laptops) seem to require Windows licence. Last I checked, Dell and HP sold desktop workstations without Linux, but the catch is that it often costs an extra $100 vs. the Windows version.