I wonder if the highly temperature-dependent power consumption can be blamed on the new digital PWM power regulation chips? Guess we'll never know for sure.
The power consumption of silicon is temperature-dependent, but it's usually small enough to be ignored. Forty watts difference is HUGE.
I owned the only graphics card to support TruForm in hardware (Radeon 8500), and I played exactly one game with TruForm support (Counterstrike), and boy was it disappointing. Will TruForm II suffer a similar fate?
I think ATI made the better move here. They have been recouping the research money on unified shader GPUs from a much bigger market segment, though it does make it appear they are lagging behind in the PC gaming sector.
You're missing one fact: the PC GPU market is MUCH LARGER market than the console GPU market.
Here are some recent sales numbers: 76 million units in Q3 2006. With ATI holding roughly 1/4 of the market (~18 million), that's more units than ATI sold in the last 6 months on the 360 and Wii combined, let-alone in the last quarter. The sad thing is, the Wii and 360 sales will likely go down from here, but the PC graphics market (overall) keeps improving with every quarter.
Furthermore, the share of discrete GPUs (where the real money is) was a massive 26 million, and those were split solely between Nvidia and ATI. You don't have to be a rocket scientist to know that ATI makes more money on an x1950 Pro or x2900 XT chipset sale than on their license fee for an Xbox 360 GPU.
Were they supposed to bring the ashes down with the rocket? I thought the whole purpose of this flight was to take the ashes up into space and scatter them.
Bring the ashes BACK? Sounds kinda bass-ackwards to me.
Agreed, you only need so-much graphics for certain games, although I will point out: you can create graphically-rich games just using the fixed-function T&L pipeline, you just don't see it these days because developers spend most of their time with the more flexible DX9/10 setup, and only include a simple T&L pipeline as an afterthought.
One great example of a pure T&L game: Battlefield 1942. All the effects are texture-driven or fixed-pipeline driven.
If you have enough memory and enough bandwidth, you can do impressive things with fixed pipelines. However, I will say I don't think Nintendo can pull this again: if the successor to the Wii doesn't have a flashy upgraded graphics chip, I don't think it will sell well.
Cell phones are great for re-organizing on-the-fly. Like say, last weekend, when we discovered en-route that the club we planned to hit was closed (expired liquor license, it happens). Thirty minutes of phone-tag in car (mind you, I wasn't driving) gave us an alternative. You can't do that kind of thing without a cell phone, and when you're dealing with multiple people it happens all the time.
Cell phones are great for emergencies, and if you think that just covers auto breakdowns, you're not being very creative. When my father died two years ago, I immediately jumped in my car at 1AM and started calling relatives at the same time (no, I don't normally do this). Who else was supposed to do it, my mother was an emotional wreck. She needed me there as soon as possible, but I also needed to let the family know as soon as possible. The cell phone also allowed family to call me back and tell me travel arrangements, regardless of where I was that night.
I can definitely recall times I wish I'd HAD a cell phone. I took a trip to London, and nobody had a working cell phone. At major sites, several times people got separated, and some folks had to spend an hour or more looking for the rest of the group. We coped with the situation, but in retrospect some throwaway cell phones would have been nice.
My girlfriend gets even more use out of her cell than I do, because she takes public transpostation to and from her college night classes. She gets bored and calls people (usually me) to while away the hour-long ride.
Cell phones are not necessary, but they give me more free time and freedom than ever.
If you buy a car, you own that car, when you buy a CD you own the physical disc
I think your problem with this whole issue is you are fettered by the fact that, unlike cars, CDs (and more specifically, the data contained therin) can last for centuries (if they are manufactured without defects, and are cared for properly).
This means the market for a device like a CD is inherently more self-limiting than a car, and because that's "not fair" in your eyes, you feel that we should artificially force the CD market to be more like that of the car market.
I'm sorry, but it doesn't work that way. Let's have ourselves a history lesson.
Way back in the day the record companies sold LPs. People built up collections on LP, which they had to replace occasionally due to wear-and-tear. Unfortunately, these replacement sales were pretty low even with a format that degraded as easily as LP, so the record companies wanted something new.
The record companies started backing CrO2 8-track and cassette tapes in the mid 70s as a successor to LPs, and this tactic largely worked. For people who couldn't afford expensive turntables and receivers, tapes sounded just as good, and the tapes were also much smaller. People replaced their libraries wth tapes for convenience.
Cassette tapes had the downside of degrading like LPs, and cassette tapes was expensive to manufacture (moving parts, lots of tape), so the record companies sold people on yet-another format to replace their collections with: the Compact Disc. By making it digital and optical, the format could both offer better dynamic range and longer lifetime than any format before it, guaranteeing it would replace tapes and encourage people to repurchase their music.
The long-term downside, something the music companies never considered, is that CDs are "good enough" all-around. They really do have more dynamic range and sample rate than most people need, and they really do last a lifetime unless you purposefully mistreat them. And with the introduction of easy copying of discs (without generational losses), now there's absolutely no reason for stamped CDs to EVER be replaced.
It doesn't matter that the music contained on the CD is "licensed." The CD comes with no written license itself, and with no written contract to limit the length of that license, it must be assumed that the license is indefinite. In the days of the LP, it was assumed the license ended once the disc became unreadable, but with today's technology lossless copies (or media shifts) are one click away, and the license can be extended forever.
The record companies knew that media replacement would only last so long when every new generation gets better than the last. Don't cry for them, they made their massive amounts of money reselling people their music on new formats, and they're still making quite a bit (on new bands, new customers and yes, replacements) despite the hole they've dug for themselves.
Most recording studios these days use, at the very least, 24bit audio at between 96-196+ khz. While I agree with you that most people won't hear a difference, audiophiles will hear a difference. My mother can't tell the difference between a hissy cassette tape and a CD, but that doesn't mean there isn't one.
The interesting point here is that online music sales could potentially supply consumers with higher quality audio than currently is available with CD.
While the MPEG4 AAC is capable of 96 KHz sample rate, the iPod is not capable of playing back this quality level. While you can certainly transcode to 48 KHz, this leaves you with quality on-par with a CD. The iPod's massive momentum can most certainly hold back a standard as flexible as AAC.
Changing the way CD's play audio would take years.
As would changing the way iPods play audio. Who's going to replace their "good enough" iPod with the new 96KHz model for almost no improvement? And what benefit does Apple get hosting larger 96 KHz tracks that nobody downloads? Isn't momentum a wonderful thing?
I actually find it funny that Apple has been pushing 128k AAC all these years as "CD quality," yet only now do they admit that it's not good enough (new 256k downloads). AAC really needs between 160k and 192k to produce a transparent copy, which is about the same as mp3, and as far as I know this has actually kept audiophiles AWAY from the iTunes store. This hasn't hurt Apple's bottom-line so-far, so why should they cater to audiophiles all of a sudden?
I wasted so many hours with this game. I picked up a copy of PC Gamer with the demo floppy for Transport Tycoon, and when I finally upgraded to enough ram to run the game, I was amazed. I got so much play out of the deluxe CD version, it was practically the only game I played for an entire year. I still pick it up to dust it off, and of course, get my friends hooked whenever possible.
Two-player TTD was amazingly fun over my college IPX network, especially since you could save the game and pick it up later.
Although, I think I'd take a copy of OpenTTD with me and try out the new features recently added.
Explanation: "tain" is derived from French slang for tin, in the 1800s. But if you look at the origin of other words I listed, they are much older, so they cannot possibly be related.
Example: Contain, Obtain, Maintain and Retain are roots of the latin word tenére, which means to hold (estimated usage around 1300 AD). But because we already have a word for "hold" (from 900 AD), the base verb tenére never made it into English, except (for example) the noun tenet.
I also wonder why we can compel, repel, impel and expel but we can't just pel.
You mean how we can retain, contain, obtain, pertain, maintain, detain, but there is no defintiion for just "tain?"
But when you see other uses of the phrase "tain" like mountain, you get the feeling that "tain" refers to any general object, and is modified by the various prefixes.
What's so bad about how Windows 2000 handles USB devices?
The only difference I've seen between 2K and XP regarding USB is write caching (enabled for 2K, disabled for XP), which allows you to remove "disks" from XP without unmounting them. Not a big deal in my experience, although I can see why it might be confusing. In all other concerns, they are identical (I used Win2K for 5 years with MANY USB devices).
You want good performance on high-speed drives? You have to turn on write caching. You want to allow for "surprise removal," you have to turn off write caching. In other words, this is hardly an "easy" design decision, and it's no surprise that XP (and not Win2k) was designed with USB flash drives in mind, because they were only becoming popular around that time.
Incidentally, the write caching in XP is NOT disabled for other removable devices (Firewire, for instance).
Actually, hasn't Intel had integrated L2 cache since the Pentium Pro?
No. The L2 cache of the Pentium Pro was on a second die. The two dies were packaged together in a somewhat similar manner to Intel's current "Quad-core" chips, except the cache attached to the brand-new back-side bus (instead of sharing the front-side bus).
The first Intel processor with integrated cache was the Mendocino-core Celeron.
While the Core 2 has its roots in the P6 microarchitecture, it is an entiely different processor design with far too many changes to be called simply a "tweak" of the P6.
The reason you have not seen in-depth articles for the Core 2 is because most "review" sites on the web are staffed by people who don't know a lick about processor architectures, and they just regurgitate Intel's presskit material without any discussion on the subject. Here is a much more in-depth article concerning the architecture.
Summary of improvements since P6:
4-wide decoding instead of 3-wide. Micro-op and Micro-op fusion (more instructions decoded). Improved branch predicition, including loop detector (from Pentium M). Support for 2 128-bit packed SSE instructions per-cycle. Three dispatch units (P6 had two ALU / FPU ports). Speculative reordering of loads. And, of course, the low-latency (from Pentium M) and shared L2 cache. x86-64 support also counts for something.
Why does it take so long to make sizeable improvements on the P6 microarchitecture? The reason is complexity: every extra decode pipe, every extra issue port, every extra ALU adds exponential complexity to the design. Intel was actualy trying to get around this problem with the Pentium 4: crank up the clock speed, and you don't have to make a more complex processor!
The failure of the Pentium 4 and the complexity of massively superscalar cores is the main reason why CPU designers are moving toward multi-core as a long-term solution: superscalar architectures as complex as Core 2 (and Barcelona) are VERY difficult to design and verify.
Yup, unfortunately this is the state of PC gaming.
You have too many fancy cases surounding crap hardware, too many companies RENAMING existing products to give them new life, and far too many derivative products; it is a confusing world out there.
The PC industry could fix this problem with banchmarks and simple grading systems, but then you have to answer the tough questions:
What happens when someone has to update the benchmark? New features usually need new benchmarks. How do users get the latest data on their system?
What happens when OEMs refuse to label their PCs with these numbers because they want to sell on style? As soon as you can attach easy numbers that MEAN something, consumers will start to do a LOT more comparison shopping, and that's not good for an industry that upsells on useless features.
The PC industry makes this even worse because they charge outrageous prices for "upgrades," so most people end up going with the base model (and discover too late just why their PC was so cheap).
In the end, PC gaming is a mess for the same reason it is a panacea: choice. Unfortunately, that is also the reason why it is at a dead-end: there are only so many people who can stomach the hoard of knowledge required to play. Me, I'm going to stick around for as long as I can, because I really can't enjoy an FPS without a mouse.
So, go play flash games (they usually work everywhere), or go play console games. Or, go learn about gaming hardware. Your choice.
Nor would server or DBMS applications require dynamic threading at the hardware level. I'm quite certain threading is already built-in to the software for such applications.
The original poster is talking about reintroducing the idea of threads into single-threaded code, which is an incredibly difficult task. This is even more complicated a task than just out-of-order execution on scalar code.
I'm hardly surprised that Intel took the easy way out on this one - the "hard" fixes for this problem are on the verge of impossible. If you can make your single-threaded processor faster, by all means do it!
Oh, hell yeah! Anyone remember the disappearing car bug (that took MULTIPLE patch versions to actually *fix*)? Sure, folks: pack that trunk with all your inventory, and watch it magically disappear. Did you put a story-critical item in that trunk? You poor fuck, you're screwed! Just one of many major bugs.
Nevermind that you had to restart the game from scratch if you wanted the new patches, because your old saves wouldn't work. You people are nuts if you think Oblivion is buggier than Fallout 2...they are both bug fests!
Exactly. Bethesda's creations are massive, and sometimes they really don't get the gameplay, balancing and interactions just right.
But that's why Oblivion and Morrowind are some of the most modded games in history - it's hard to please every single player in a game world so vast, and even if it were possible there's really not enough development time.
I look at it this way: if Fallout 3 has crappy gameplay, I'll wait 6 months for the equivilant of Oscuro's Oblivion Overhaul to be released, and then enjoy a vast game with excellent balancing. At least with Bethesda at the helm you're guaranteed a good basis for modders to start with.
Face it, Fallout could be at the mercy of MUCH WORSE dev teams. I actually enjoyed Oblivion with OOO.
The Reality of the situation is the X3000 gets trounced in mnay games by the 2-year old GeForce 6150, and to get respectable (> 20fps) framerates in recent games you have to pair it up with a beefy Core 2 Duo.
Getting the picture yet? Yuk yuk. The bottom line is that you get radically better performance out of H.264 than MPEG-2 at similar bitrates. So a ~45 minute TV episode weighing in at 400MB for a total combined audio/video bitrate of around 1250 kbps gets nearly identical quality to a 2500 kbps MPEG-2 bitstream. Of course on DVD you get goodies like the 5.1 surround audio track, so it's still a better deal, but Apple's done a lot to close the gap.
I agree that H.264 can cut your required bandwidth by half to acheive the same quality, but you've missed a key fact in your numbers: 2.5 Mbps MPEG2 (standard-def) looks like crap.
The typical video bitrate of most DVDs is 4-5 Mbps. Most action movies hit the magic 6 Mbps, and most Superbit editions target 7 Mbps! H.264 is amazing, but it cannot make-up for a 4:1, let-alone a 6:1 bitrate ratio. What I've perrsonally seen backs this up: H.264 standard-def content doesn't start looking REALLY good until you approach 2 Mbps.
Here's an example of how important bitrate is: a comparison of Crouching Tiger's standard release versus Superbit. Even thought the original releases uses the DVD Forum recommended video bitrate of around 5 Mbps, the quality of the original pales in comparison to the Superbit (around 7Mbps). There are other movies listed on the same page to give you an idea of the average bitrate of GOOD action transfers to DVD.
If you really think H.264 can work miracles over MPEG2 with just 1/6th of that bitrate, I want to try whatever you're smoking, because it's obviously good shit:D
Slashdot Mirror
I wonder if the highly temperature-dependent power consumption can be blamed on the new digital PWM power regulation chips? Guess we'll never know for sure.
The power consumption of silicon is temperature-dependent, but it's usually small enough to be ignored. Forty watts difference is HUGE.
New programmable Tessellation technology
Am I the only one who reads this as TruForm II?
I owned the only graphics card to support TruForm in hardware (Radeon 8500), and I played exactly one game with TruForm support (Counterstrike), and boy was it disappointing. Will TruForm II suffer a similar fate?
I think ATI made the better move here. They have been recouping the research money on unified shader GPUs from a much bigger market segment, though it does make it appear they are lagging behind in the PC gaming sector.
You're missing one fact: the PC GPU market is MUCH LARGER market than the console GPU market.
Here are some recent sales numbers: 76 million units in Q3 2006. With ATI holding roughly 1/4 of the market (~18 million), that's more units than ATI sold in the last 6 months on the 360 and Wii combined, let-alone in the last quarter. The sad thing is, the Wii and 360 sales will likely go down from here, but the PC graphics market (overall) keeps improving with every quarter.
Furthermore, the share of discrete GPUs (where the real money is) was a massive 26 million, and those were split solely between Nvidia and ATI. You don't have to be a rocket scientist to know that ATI makes more money on an x1950 Pro or x2900 XT chipset sale than on their license fee for an Xbox 360 GPU.
Were they supposed to bring the ashes down with the rocket? I thought the whole purpose of this flight was to take the ashes up into space and scatter them.
Bring the ashes BACK? Sounds kinda bass-ackwards to me.
Agreed, you only need so-much graphics for certain games, although I will point out: you can create graphically-rich games just using the fixed-function T&L pipeline, you just don't see it these days because developers spend most of their time with the more flexible DX9/10 setup, and only include a simple T&L pipeline as an afterthought.
One great example of a pure T&L game: Battlefield 1942. All the effects are texture-driven or fixed-pipeline driven.
If you have enough memory and enough bandwidth, you can do impressive things with fixed pipelines. However, I will say I don't think Nintendo can pull this again: if the successor to the Wii doesn't have a flashy upgraded graphics chip, I don't think it will sell well.
Yes, but does it have four asses?
Cell phones are great for re-organizing on-the-fly. Like say, last weekend, when we discovered en-route that the club we planned to hit was closed (expired liquor license, it happens). Thirty minutes of phone-tag in car (mind you, I wasn't driving) gave us an alternative. You can't do that kind of thing without a cell phone, and when you're dealing with multiple people it happens all the time.
Cell phones are great for emergencies, and if you think that just covers auto breakdowns, you're not being very creative. When my father died two years ago, I immediately jumped in my car at 1AM and started calling relatives at the same time (no, I don't normally do this). Who else was supposed to do it, my mother was an emotional wreck. She needed me there as soon as possible, but I also needed to let the family know as soon as possible. The cell phone also allowed family to call me back and tell me travel arrangements, regardless of where I was that night.
I can definitely recall times I wish I'd HAD a cell phone. I took a trip to London, and nobody had a working cell phone. At major sites, several times people got separated, and some folks had to spend an hour or more looking for the rest of the group. We coped with the situation, but in retrospect some throwaway cell phones would have been nice.
My girlfriend gets even more use out of her cell than I do, because she takes public transpostation to and from her college night classes. She gets bored and calls people (usually me) to while away the hour-long ride.
Cell phones are not necessary, but they give me more free time and freedom than ever.
If you buy a car, you own that car, when you buy a CD you own the physical disc
I think your problem with this whole issue is you are fettered by the fact that, unlike cars, CDs (and more specifically, the data contained therin) can last for centuries (if they are manufactured without defects, and are cared for properly).
This means the market for a device like a CD is inherently more self-limiting than a car, and because that's "not fair" in your eyes, you feel that we should artificially force the CD market to be more like that of the car market.
I'm sorry, but it doesn't work that way. Let's have ourselves a history lesson.
Way back in the day the record companies sold LPs. People built up collections on LP, which they had to replace occasionally due to wear-and-tear. Unfortunately, these replacement sales were pretty low even with a format that degraded as easily as LP, so the record companies wanted something new.
The record companies started backing CrO2 8-track and cassette tapes in the mid 70s as a successor to LPs, and this tactic largely worked. For people who couldn't afford expensive turntables and receivers, tapes sounded just as good, and the tapes were also much smaller. People replaced their libraries wth tapes for convenience.
Cassette tapes had the downside of degrading like LPs, and cassette tapes was expensive to manufacture (moving parts, lots of tape), so the record companies sold people on yet-another format to replace their collections with: the Compact Disc. By making it digital and optical, the format could both offer better dynamic range and longer lifetime than any format before it, guaranteeing it would replace tapes and encourage people to repurchase their music.
The long-term downside, something the music companies never considered, is that CDs are "good enough" all-around. They really do have more dynamic range and sample rate than most people need, and they really do last a lifetime unless you purposefully mistreat them. And with the introduction of easy copying of discs (without generational losses), now there's absolutely no reason for stamped CDs to EVER be replaced.
It doesn't matter that the music contained on the CD is "licensed." The CD comes with no written license itself, and with no written contract to limit the length of that license, it must be assumed that the license is indefinite. In the days of the LP, it was assumed the license ended once the disc became unreadable, but with today's technology lossless copies (or media shifts) are one click away, and the license can be extended forever.
The record companies knew that media replacement would only last so long when every new generation gets better than the last. Don't cry for them, they made their massive amounts of money reselling people their music on new formats, and they're still making quite a bit (on new bands, new customers and yes, replacements) despite the hole they've dug for themselves.
Most recording studios these days use, at the very least, 24bit audio at between 96-196+ khz. While I agree with you that most people won't hear a difference, audiophiles will hear a difference. My mother can't tell the difference between a hissy cassette tape and a CD, but that doesn't mean there isn't one.
The interesting point here is that online music sales could potentially supply consumers with higher quality audio than currently is available with CD.
While the MPEG4 AAC is capable of 96 KHz sample rate, the iPod is not capable of playing back this quality level. While you can certainly transcode to 48 KHz, this leaves you with quality on-par with a CD. The iPod's massive momentum can most certainly hold back a standard as flexible as AAC.
Changing the way CD's play audio would take years.
As would changing the way iPods play audio. Who's going to replace their "good enough" iPod with the new 96KHz model for almost no improvement? And what benefit does Apple get hosting larger 96 KHz tracks that nobody downloads? Isn't momentum a wonderful thing?
I actually find it funny that Apple has been pushing 128k AAC all these years as "CD quality," yet only now do they admit that it's not good enough (new 256k downloads). AAC really needs between 160k and 192k to produce a transparent copy, which is about the same as mp3, and as far as I know this has actually kept audiophiles AWAY from the iTunes store. This hasn't hurt Apple's bottom-line so-far, so why should they cater to audiophiles all of a sudden?
I wasted so many hours with this game. I picked up a copy of PC Gamer with the demo floppy for Transport Tycoon, and when I finally upgraded to enough ram to run the game, I was amazed. I got so much play out of the deluxe CD version, it was practically the only game I played for an entire year. I still pick it up to dust it off, and of course, get my friends hooked whenever possible.
Two-player TTD was amazingly fun over my college IPX network, especially since you could save the game and pick it up later.
Although, I think I'd take a copy of OpenTTD with me and try out the new features recently added.
Two things we sill cannot model %100 correctly, let alone produce something better:
1. spider silk.
2. insect flight.
It's a joke. Woosh.
Explanation: "tain" is derived from French slang for tin, in the 1800s. But if you look at the origin of other words I listed, they are much older, so they cannot possibly be related.
Example: Contain, Obtain, Maintain and Retain are roots of the latin word tenére, which means to hold
(estimated usage around 1300 AD). But because we already have a word for "hold" (from 900 AD), the base verb tenére never made it into English, except (for example) the noun tenet.
And hence, we have no tain.
I also wonder why we can compel, repel, impel and expel but we can't just pel.
You mean how we can retain, contain, obtain, pertain, maintain, detain, but there is no defintiion for just "tain?"
But when you see other uses of the phrase "tain" like mountain, you get the feeling that "tain" refers to any general object, and is modified by the various prefixes.
Welcome to English.
On the PC side of the world, we've been enjoying various 3rd-party Blu-Ray drives and various third-party Blu-Ray video players for a year now.
As a Mac user, I'm rather disappiointed. But that's why I'm also a PC user - it helps me avoid disappointment when Apple decides to sit on the fence.
You know, we wouldn't be having this problem with high-G if we had picked Bulrathi instead of Humans in the first place!
What's so bad about how Windows 2000 handles USB devices?
The only difference I've seen between 2K and XP regarding USB is write caching (enabled for 2K, disabled for XP), which allows you to remove "disks" from XP without unmounting them. Not a big deal in my experience, although I can see why it might be confusing. In all other concerns, they are identical (I used Win2K for 5 years with MANY USB devices).
You want good performance on high-speed drives? You have to turn on write caching. You want to allow for "surprise removal," you have to turn off write caching. In other words, this is hardly an "easy" design decision, and it's no surprise that XP (and not Win2k) was designed with USB flash drives in mind, because they were only becoming popular around that time.
Incidentally, the write caching in XP is NOT disabled for other removable devices (Firewire, for instance).
Actually, hasn't Intel had integrated L2 cache since the Pentium Pro?
No. The L2 cache of the Pentium Pro was on a second die. The two dies were packaged together in a somewhat similar manner to Intel's current "Quad-core" chips, except the cache attached to the brand-new back-side bus (instead of sharing the front-side bus).
The first Intel processor with integrated cache was the Mendocino-core Celeron.
While the Core 2 has its roots in the P6 microarchitecture, it is an entiely different processor design with far too many changes to be called simply a "tweak" of the P6.
The reason you have not seen in-depth articles for the Core 2 is because most "review" sites on the web are staffed by people who don't know a lick about processor architectures, and they just regurgitate Intel's presskit material without any discussion on the subject. Here is a much more in-depth article concerning the architecture.
Summary of improvements since P6:
4-wide decoding instead of 3-wide.
Micro-op and Micro-op fusion (more instructions decoded).
Improved branch predicition, including loop detector (from Pentium M).
Support for 2 128-bit packed SSE instructions per-cycle.
Three dispatch units (P6 had two ALU / FPU ports).
Speculative reordering of loads.
And, of course, the low-latency (from Pentium M) and shared L2 cache. x86-64 support also counts for something.
Why does it take so long to make sizeable improvements on the P6 microarchitecture? The reason is complexity: every extra decode pipe, every extra issue port, every extra ALU adds exponential complexity to the design. Intel was actualy trying to get around this problem with the Pentium 4: crank up the clock speed, and you don't have to make a more complex processor!
The failure of the Pentium 4 and the complexity of massively superscalar cores is the main reason why CPU designers are moving toward multi-core as a long-term solution: superscalar architectures as complex as Core 2 (and Barcelona) are VERY difficult to design and verify.
Yup, unfortunately this is the state of PC gaming.
You have too many fancy cases surounding crap hardware, too many companies RENAMING existing products to give them new life, and far too many derivative products; it is a confusing world out there.
The PC industry could fix this problem with banchmarks and simple grading systems, but then you have to answer the tough questions:
What happens when someone has to update the benchmark? New features usually need new benchmarks. How do users get the latest data on their system?
What happens when OEMs refuse to label their PCs with these numbers because they want to sell on style? As soon as you can attach easy numbers that MEAN something, consumers will start to do a LOT more comparison shopping, and that's not good for an industry that upsells on useless features.
The PC industry makes this even worse because they charge outrageous prices for "upgrades," so most people end up going with the base model (and discover too late just why their PC was so cheap).
In the end, PC gaming is a mess for the same reason it is a panacea: choice. Unfortunately, that is also the reason why it is at a dead-end: there are only so many people who can stomach the hoard of knowledge required to play. Me, I'm going to stick around for as long as I can, because I really can't enjoy an FPS without a mouse.
So, go play flash games (they usually work everywhere), or go play console games. Or, go learn about gaming hardware. Your choice.
Nor would server or DBMS applications require dynamic threading at the hardware level. I'm quite certain threading is already built-in to the software for such applications.
The original poster is talking about reintroducing the idea of threads into single-threaded code, which is an incredibly difficult task. This is even more complicated a task than just out-of-order execution on scalar code.
I'm hardly surprised that Intel took the easy way out on this one - the "hard" fixes for this problem are on the verge of impossible. If you can make your single-threaded processor faster, by all means do it!
Oh, hell yeah! Anyone remember the disappearing car bug (that took MULTIPLE patch versions to actually *fix*)? Sure, folks: pack that trunk with all your inventory, and watch it magically disappear. Did you put a story-critical item in that trunk? You poor fuck, you're screwed! Just one of many major bugs.
Nevermind that you had to restart the game from scratch if you wanted the new patches, because your old saves wouldn't work. You people are nuts if you think Oblivion is buggier than Fallout 2...they are both bug fests!
Exactly. Bethesda's creations are massive, and sometimes they really don't get the gameplay, balancing and interactions just right.
But that's why Oblivion and Morrowind are some of the most modded games in history - it's hard to please every single player in a game world so vast, and even if it were possible there's really not enough development time.
I look at it this way: if Fallout 3 has crappy gameplay, I'll wait 6 months for the equivilant of Oscuro's Oblivion Overhaul to be released, and then enjoy a vast game with excellent balancing. At least with Bethesda at the helm you're guaranteed a good basis for modders to start with.
Face it, Fallout could be at the mercy of MUCH WORSE dev teams. I actually enjoyed Oblivion with OOO.
"Target" and "Reality" are two different things.
The "Target" of the G965 is to raise performance above anything available today.
The "Reality" of the situation is: the drivers are terrible. Vertex shading is still done in-software, just like the G950. As-of March (six months after release), there are still no updates available, and the rumored "beta" drivers are still just that - a rumor.
The Reality of the situation is the X3000 gets trounced in mnay games by the 2-year old GeForce 6150, and to get respectable (> 20fps) framerates in recent games you have to pair it up with a beefy Core 2 Duo.
"I have lived too long. Hi ho."
Getting the picture yet? Yuk yuk. The bottom line is that you get radically better performance out of H.264 than MPEG-2 at similar bitrates. So a ~45 minute TV episode weighing in at 400MB for a total combined audio/video bitrate of around 1250 kbps gets nearly identical quality to a 2500 kbps MPEG-2 bitstream. Of course on DVD you get goodies like the 5.1 surround audio track, so it's still a better deal, but Apple's done a lot to close the gap.
:D
I agree that H.264 can cut your required bandwidth by half to acheive the same quality, but you've missed a key fact in your numbers: 2.5 Mbps MPEG2 (standard-def) looks like crap.
The typical video bitrate of most DVDs is 4-5 Mbps. Most action movies hit the magic 6 Mbps, and most Superbit editions target 7 Mbps! H.264 is amazing, but it cannot make-up for a 4:1, let-alone a 6:1 bitrate ratio. What I've perrsonally seen backs this up: H.264 standard-def content doesn't start looking REALLY good until you approach 2 Mbps.
Here's an example of how important bitrate is: a comparison of Crouching Tiger's standard release versus Superbit. Even thought the original releases uses the DVD Forum recommended video bitrate of around 5 Mbps, the quality of the original pales in comparison to the Superbit (around 7Mbps). There are other movies listed on the same page to give you an idea of the average bitrate of GOOD action transfers to DVD.
If you really think H.264 can work miracles over MPEG2 with just 1/6th of that bitrate, I want to try whatever you're smoking, because it's obviously good shit