Well, they ARE generally cheap POS devices. There are a few exceptions but I'd advise against buying the cheapest or first item available with XYZ functionality anyway.
I think the "big guys" are trying to refrain from adding too much too quickly, letting the small ones add features that the RIAA/MPAA might bitch about (DIVX/MP3) so that they can serve as lightning rods.
I think they have a very good size market share simply on affordability, it is easier to convince maybe five people to buy $50 player vs. one person buying a $100 player. They also use cheaper decoders and video amps, but the difference may not be apparent on the cheap TVs most people own.
"we"? Are you on a "side"? I'd guess that you have nothing to do with any manufacturer or group involved with making new computer technology. I always thought it odd to take sides in something that has nothing to do with the person.
But I digress.
I think it just might be economies of scale. Billions of dollars are spent in CPU design and improved manufacturing to make a cheaper product, and to make that cheap product better. When one can ammortize the cost over millions of units one can fund a lot of research.
Using reference board pricing isn't a good example as reference boards are always expensive, in part because they are designed and supported as a developer product.
The x86 reference boards are even slightly more expensive:
http://www.compulab.co.il/price-eval-kits.htm
That said, economies of scale favor x86, so non-x86 mass production boards will simply be expensive. Economies of scale are a b!tch. That leaves it to commercial uses because the boards just won't run consumer software, thus hopefull preventing component theft. I made a battery powered box that could run for years on a battery, rather than choosing a standard 9V or several "AA" cells, they went to "B" cells so people won't steal the batteries!
GCC sucks on everything. Really. It performs poorly compared to Intel's compilers, Digital's compilers, IBM compilers, etc. I think it might have done better than SUN compilers though, but I heard that's not saying much. GCC's only strong points are GPL licensing and possibly the broadest hardware support, and stable output I guess, it was everything but fast.
DEC just happened to release a special compiler for Linux, GEM, I think. End users could get it, I have a downloaded beta of it too. The back end is the same as on Tru64's, only the Linux front end was considered beta.
The Linux engineers in DEC where having a hard time convincing the GCC group to improve the front-end to back-end communication because they could optimize better, but they refused. Remember, the GCC group was the bunch of miscreants that effectively required the EGCS split to get anything done even for just Pentium optimizations!
The SSC was too controversial. I think there was some massive uproar once people got wind of how the contractors were wasting money and misusing the property. I wish I remembered the details, I'll have to look it up again.
""We've used the default factory settings for all projectors so we get a fair comparison between them.""
"Perhaps for some measurements they used defaults and some they adjusted to match the reference image, but that's far from clear."
Oh dear. This is a huge mistake. Every display regardless of type MUST be tuned for its setting. There are slight variations between projectors, and settings ideal for one room and lighting suck for other rooms and lightings, so they just pick a generic setting and make that the stock "factory settings". One does NOT do default settings if you want the best picture or a fair comparison of what they can do.
The bulb cost is high but I think it is worth it. Bulbs cost between $300-$500 but they last between 1000 and 5000 hours depending on the model of projector you bought. If you assume you watch two hours of video per evening, every evening, that is nearly 1.4 years for every 1000 hours. So if you get a unit with a 3000 hour bulb rating, watch two hours a night, then that means four years of use before replacing it.
On a 2000 hour bulb, if you pay the SRP of $500, that is twenty five cents per hour. That is a pittance compared to the typical geek daily Mountain Dew budget, although I don't drink any.
I haven't done this yet, but a change jar might be a good way to save up for the next bulb.
I keep a smaller TV as well, mostly for watching regular TV and tapes.
Even with "full range" speakers, a subwoofer still helps, and they are even used in most movie theaters. Most full range speakers corner at between 30 and 40 hz. A sub can improve the response between the 15-30Hz range. In terms of money spent, getting a full range speaker that corners at 15Hz is probably prohibitivly expensive compared to just getting a sub that "fills the gap".
I will have to disagree on this point. Digital transportation of an audio signal is no different than digital transportation of an IP packet. The data in should be data out. It doesn't matter if it is 75 ohm audio cable, optical signal or CAT5.
People have done their own tests with TOSLINK and coax connections - any decent cable will transport billions of bits with only a _few_ bit errors. A $5 optical cable generally perform as well as a $100 optical cable in this regard. There should be no difference between optical and coaxial signals, if there are, then there is often something wrong with the sending end or the recieving end being out of spec.
Also, one can assemble a pretty decent six channel sound system for about that price. As it is five speakers in ONE cabinet, I think I'd trade the tricky electronics for four more cabinets and an amp, thank you.
Yeah, it sucks that NASA has poor management and a shrinking budget. They've even tried to suck up to at least one presidential administration with "Goresat" which had no real scientific merit, and that IIRC, Gore had argueed and voted to reduce NASA budget seven out of eight times.
People that expect technological leaps in bounds in simple chemical reactions to the scale of Moore's law simply haven't been paying attention in science class.
They should be asking why such good and advanced technology is sucking down more and more power at every iteration. I think the reason is that people want performance first and battery life is an afterthought. With a market like that, then that is your reason why each new generation can do more but requires more power. To take less power for each generation, the market needs to demand longer battery life first or balance battery life better with performance. More can still be done with less power if we let it.
Actually, people have pointed out that, when adjusted for inflation over the years (two decades now?), the target prices for CDs originally promoted by the RIAA has been reached. I wish I had a source on that.
Of course, one other argument I've complained about is people were complaining about $20 CDs. I never did see them until I went into a mall CD store, and I was shocked to see they charged it, but it's people's own fault for paying full S.R.P.
So far the only disagreeable zealotry I've found was one guy the pro Wintel side where many counterexamples were returned with profanity. I often wish computer zealots would get a grip, but man that one needed a padded cell.
I think Matrox has plenty of relevance, just not so much in the gaming arena.
I will be buying another Matrox, either a 650, 750 or a Parhelia for my HTPC. I love the 450 but given its age, it doesn't have DxVA / IDCT acceleration and the overlay scaler isn't so great, it has been supersceeded in quality by newer designs from Matrox and their competitors. The overlay scaler was still better than the original Radeon though. I want a new Matrox for Dual Head Clone so I can run either or both of two displays (21" CRT and LCD presentation projector) and have the same image on both, without an image degrading splitter box.
Also, the people that formed 3DFX came out of a Media Vision video card project the Pro Graphics 1024. Back in the VLB days, this card ran 24 bits depth (standard!) at XGA resolution and I could play half a dozen video clips simultaneously and not really feel like slowing down an AMD DX4-100. Media Vision was in financial trouble so that project fell by the wayside and they had resurfaced as an audio chip and codec maker.
I don't think legacy hardware is a problem at all. That market isn't like the PC market where the maker pretends that it didn't exist if it is more than three years old.
If you check out Sun, you'll find that they still sell parts for even their oldest systems. That and more is what the mainframe market is about.
While one can play games on a Matrox, they were NEVER the best for any 3D games, so I wouldn't use gaming performance as a reason that they are going "down the tubes". While games are a significant part of the graphic card segment, they are hardly the only part, and because the company really doesn't compete in games doesn't mean they don't have a strong market niche.
For everything else that isn't 3D gaming, I've found Matrox cards to do plenty well and they did everything I wanted, plus a few things their competitors didn't do at all.
They do have some specialized cards too, such as four-head and a card with outputs designed for 12 megapixel displays. They also have some heavy duty cards designed for professional real time multi-stream video editing.
Slashdot Mirror
Well, they ARE generally cheap POS devices. There are a few exceptions but I'd advise against buying the cheapest or first item available with XYZ functionality anyway.
I think the "big guys" are trying to refrain from adding too much too quickly, letting the small ones add features that the RIAA/MPAA might bitch about (DIVX/MP3) so that they can serve as lightning rods.
I think they have a very good size market share simply on affordability, it is easier to convince maybe five people to buy $50 player vs. one person buying a $100 player. They also use cheaper decoders and video amps, but the difference may not be apparent on the cheap TVs most people own.
have we simply surpassed them?
"we"? Are you on a "side"? I'd guess that you have nothing to do with any manufacturer or group involved with making new computer technology. I always thought it odd to take sides in something that has nothing to do with the person.
But I digress.
I think it just might be economies of scale. Billions of dollars are spent in CPU design and improved manufacturing to make a cheaper product, and to make that cheap product better. When one can ammortize the cost over millions of units one can fund a lot of research.
Using reference board pricing isn't a good example as reference boards are always expensive, in part because they are designed and supported as a developer product.
The x86 reference boards are even slightly more expensive:
http://www.compulab.co.il/price-eval-kits.htm
That said, economies of scale favor x86, so non-x86 mass production boards will simply be expensive. Economies of scale are a b!tch. That leaves it to commercial uses because the boards just won't run consumer software, thus hopefull preventing component theft. I made a battery powered box that could run for years on a battery, rather than choosing a standard 9V or several "AA" cells, they went to "B" cells so people won't steal the batteries!
In three years, no one you know will still be at the same email addresses, so it's a good time to update those addresses anyway!
GCC sucks on everything. Really. It performs poorly compared to Intel's compilers, Digital's compilers, IBM compilers, etc. I think it might have done better than SUN compilers though, but I heard that's not saying much. GCC's only strong points are GPL licensing and possibly the broadest hardware support, and stable output I guess, it was everything but fast.
DEC just happened to release a special compiler for Linux, GEM, I think. End users could get it, I have a downloaded beta of it too. The back end is the same as on Tru64's, only the Linux front end was considered beta.
The Linux engineers in DEC where having a hard time convincing the GCC group to improve the front-end to back-end communication because they could optimize better, but they refused. Remember, the GCC group was the bunch of miscreants that effectively required the EGCS split to get anything done even for just Pentium optimizations!
Nah, the conspiracy theorists would have used it to claim that Bush Jr. wants to brainwash everyone.
The SSC was too controversial. I think there was some massive uproar once people got wind of how the contractors were wasting money and misusing the property. I wish I remembered the details, I'll have to look it up again.
It could be a small heater. I think the 603e ran on only a couple watts.
""We've used the default factory settings for all projectors so we get a fair comparison between them.""
"Perhaps for some measurements they used defaults and some they adjusted to match the reference image, but that's far from clear."
Oh dear. This is a huge mistake. Every display regardless of type MUST be tuned for its setting. There are slight variations between projectors, and settings ideal for one room and lighting suck for other rooms and lightings, so they just pick a generic setting and make that the stock "factory settings". One does NOT do default settings if you want the best picture or a fair comparison of what they can do.
The bulb cost is high but I think it is worth it. Bulbs cost between $300-$500 but they last between 1000 and 5000 hours depending on the model of projector you bought. If you assume you watch two hours of video per evening, every evening, that is nearly 1.4 years for every 1000 hours. So if you get a unit with a 3000 hour bulb rating, watch two hours a night, then that means four years of use before replacing it.
On a 2000 hour bulb, if you pay the SRP of $500, that is twenty five cents per hour. That is a pittance compared to the typical geek daily Mountain Dew budget, although I don't drink any.
I haven't done this yet, but a change jar might be a good way to save up for the next bulb.
I keep a smaller TV as well, mostly for watching regular TV and tapes.
Even with "full range" speakers, a subwoofer still helps, and they are even used in most movie theaters. Most full range speakers corner at between 30 and 40 hz. A sub can improve the response between the 15-30Hz range. In terms of money spent, getting a full range speaker that corners at 15Hz is probably prohibitivly expensive compared to just getting a sub that "fills the gap".
I will have to disagree on this point. Digital transportation of an audio signal is no different than digital transportation of an IP packet. The data in should be data out. It doesn't matter if it is 75 ohm audio cable, optical signal or CAT5.
People have done their own tests with TOSLINK and coax connections - any decent cable will transport billions of bits with only a _few_ bit errors. A $5 optical cable generally perform as well as a $100 optical cable in this regard. There should be no difference between optical and coaxial signals, if there are, then there is often something wrong with the sending end or the recieving end being out of spec.
Also, one can assemble a pretty decent six channel sound system for about that price. As it is five speakers in ONE cabinet, I think I'd trade the tricky electronics for four more cabinets and an amp, thank you.
Yeah, it sucks that NASA has poor management and a shrinking budget. They've even tried to suck up to at least one presidential administration with "Goresat" which had no real scientific merit, and that IIRC, Gore had argueed and voted to reduce NASA budget seven out of eight times.
People that expect technological leaps in bounds in simple chemical reactions to the scale of Moore's law simply haven't been paying attention in science class.
They should be asking why such good and advanced technology is sucking down more and more power at every iteration. I think the reason is that people want performance first and battery life is an afterthought. With a market like that, then that is your reason why each new generation can do more but requires more power. To take less power for each generation, the market needs to demand longer battery life first or balance battery life better with performance. More can still be done with less power if we let it.
At 100 watts per CPU being fairly high end today, an automobile with 200kWatt engines (~270HP) are a FAR greater concern.
Translation:
I can build a bad-ass server, yet I can't break things into paragraphs to save my life!
I really don't see why this thing has sprouted so many hecklers. If you don't like it, don't buy it.
The Slashdot readership has often been a bunch of intellectual parrots but it has gone to a new low.
You are right. They are probably unbalanced gaming twits. Ignore them.
Few to no real servers actually need 3D, and 8MB is often judged to be plenty enough if you look at the boards designed for server use.
The only exception is if these people are making their every day system into a server, which may not be advisable for anything.
Actually, people have pointed out that, when adjusted for inflation over the years (two decades now?), the target prices for CDs originally promoted by the RIAA has been reached. I wish I had a source on that.
Of course, one other argument I've complained about is people were complaining about $20 CDs. I never did see them until I went into a mall CD store, and I was shocked to see they charged it, but it's people's own fault for paying full S.R.P.
I see that being a possibility.
So far the only disagreeable zealotry I've found was one guy the pro Wintel side where many counterexamples were returned with profanity. I often wish computer zealots would get a grip, but man that one needed a padded cell.
I think Matrox has plenty of relevance, just not so much in the gaming arena.
I will be buying another Matrox, either a 650, 750 or a Parhelia for my HTPC. I love the 450 but given its age, it doesn't have DxVA / IDCT acceleration and the overlay scaler isn't so great, it has been supersceeded in quality by newer designs from Matrox and their competitors. The overlay scaler was still better than the original Radeon though. I want a new Matrox for Dual Head Clone so I can run either or both of two displays (21" CRT and LCD presentation projector) and have the same image on both, without an image degrading splitter box.
Also, the people that formed 3DFX came out of a Media Vision video card project the Pro Graphics 1024. Back in the VLB days, this card ran 24 bits depth (standard!) at XGA resolution and I could play half a dozen video clips simultaneously and not really feel like slowing down an AMD DX4-100. Media Vision was in financial trouble so that project fell by the wayside and they had resurfaced as an audio chip and codec maker.
How much effort did it take for you to get your system to that point?
When a person wants to install another program? Install new hardware?
I don't think legacy hardware is a problem at all. That market isn't like the PC market where the maker pretends that it didn't exist if it is more than three years old.
If you check out Sun, you'll find that they still sell parts for even their oldest systems. That and more is what the mainframe market is about.
While one can play games on a Matrox, they were NEVER the best for any 3D games, so I wouldn't use gaming performance as a reason that they are going "down the tubes". While games are a significant part of the graphic card segment, they are hardly the only part, and because the company really doesn't compete in games doesn't mean they don't have a strong market niche.
For everything else that isn't 3D gaming, I've found Matrox cards to do plenty well and they did everything I wanted, plus a few things their competitors didn't do at all.
They do have some specialized cards too, such as four-head and a card with outputs designed for 12 megapixel displays. They also have some heavy duty cards designed for professional real time multi-stream video editing.