Err, IBM didn't just license Cyrix's chips, they MADE the chips for Cyrix! After a while they started keeping a few of the chips that they made and sold them under their own brand name, though I really don't think that they undercut them in price. In fact, I'm pretty certain that the IBM branded Cyrix chips were typically more expensive than the Cyrix branded ones, even though they were identical (one minor exception, I think they had slightly different core clock speed and bus speeds/clock multipliers on a few models).
What killed Cyrix had nothing to do with IBM, and everything to do with the fact that they didn't have enough money to make it as an independant company, and then they were sold to National Semiconductor in what has to have been one of the most badly botched sales in microprocessor history. Both companies ended up suffering badly as a result. NS finally gave up on the money-pit named Cyrix and essentially sold the name to VIA (the whole company was sold, but VIA than proceeded to sack pretty much the entire Cyrix team and ditched all their processor designs, than slapped the Cyrix name on the IDT Winchip processor design they had purchased at about the same time). Mind you, VIA has since dropped even the Cyrix name.
So, long story short, a lot of bad business deals and mismanagement killed Cyrix, not IBM.
If you had a K6-350 that was beaten by a P166, then you had a computer that was improperly setup, or a REALLY odd-ball setup.
Even the "really weak" floating point unit of the K6 was only about 10-20% slower than the PentiumMMX in all except the most extreme situations (and, on the other end of things, there were some floating point applications that ran faster on the K6). This is especially true by the time they got to the K6-2 350MHz chip, which had a 100MHz bus speed and a slightly improved floating point unit as compared to the original K6.
The K6 had it's flaws here and there, but for the most part it performed quite well. The floating point unit wasn't pipelined, and usually took two clock cycles per instruction, while the PentiumMMX was pipelined and therefore could sometimes pump out instructions in one clock cycle, however this didn't result in all that huge of a performance difference like some people have suggested.
Actually, the original poster was correct, at least sort of.
AMD did produce 286, 386 and 486 processors that were faster than any comperable chips that Intel ever produce (16MHz 286's if my memory serves, 40MHz 386s and 133MHz 486s). However, they did so after Intel had already released their next generation of chips.
Still, AMD did have the speed advantage at one point in time. When AMD brought out their 386DX40, the fastest chips that Intel had was their 486DX25, and AMD's 40MHz 386 was almost always faster than Intel's 25MHz 486, even though the 486 was faster clock for clock. Intel soon bumped the 486 up to 33MHz, and than it was more or less toss-up between the two, though AMD gained a lot of market share since their chips and the supporting motherboards were a LOT cheaper.
The 233MHz K6 was also, on average, marginally faster than the fastest Intel chips (the PentiumMMX 200 and PPro 200MHz) when it was released. However, that point became rather moot since it took AMD about 4 months to produce more than trivial quantities of K6's at that speed grade, and Intel brought out their PII at 233-300MHz only a few weeks after the K6 was first released.
All well and good in concept, except for that fact that there isn't a single operating system, or even a compiler in existance for Transmeta's native VLIW mode. Actually, for that matter, there isn't even any documentation available for it's VLIW instructions, so even assembly or raw byte code is out.
Face it, the Transmeta Crusoe IS an x86 chip, it just uses a really odd-ball way of getting around any Intel patents. Also, with Intel's Banias processor on the near horizon, Transmeta's days are severely numbered IMO.
Heat spreaders aren't really that great when it comes to actually making the chip run cooler, all they do is exactly what their name suggests, they spread heat.
The reason why this is a good thing is that modern processors do not generate their heat evenly, so you might have the bottom left 25% of the chip generating 50W, while the other 75% of the chip is mostly idle, only generating a few watts here or there. This is a bad thing for a variety of reasons, beyond the obvious fact that one area has to be cooled more than others. When it comes to actually dissipating the heat though, what you want is to get this heat into the furthest corners of the heatsink as quickly as possible. This is why copper inlays are a good thing for heatsinks (they spread the heat away from the hot core towards the edges of the heatsink faster than aluminium does), however even aluminium conducts heat a lot better than the P4's heat spreaders do.
This sort of thing is, presumably, less of a problem for the Athlon as compared to the P4 simply due to die sizes. The P4 is a MUCH larger die (nearly twice the size) when compared to the Athlon. So, while the distance between the hottest and coldest points in an Athlon are always going to be quite small, they can be relatively large on a P4.
As for fixing the heatsink to the motherboard instead of the socket, this is a damn good thing if you ask me. AMD originally thought that this would be how things would be done with the Athlon, and specified 4 holes around the socket to do just such a thing. You can even buy heatsinks that attach this way (my Athlon is cooled by just such a heatsink, an Alpha PAL 8045), however the vast majority of heatsink designers just flat out ignored these holes and just attached to the heatsink. Intel was a bit more forcefull, and absultely required ALL heatsinks to attached either to the motherboard or even to bolt to the case itself (the first P4s, those that came in a socket 423 package, required a special case that the heatsink bolted on to). Intel's currently design of using clips to clamp the heatsink down to the motherboard is a fairly good one IMO, though it does put an awful lot of strain on the board. AMD's plan for the Hammer/Opteron is similar, and it looks like they're going to try to avoid some of the motherboard bending/strain, however it remains to be seen just how well it will work in practice.
In any case, it remains to be seen just what the power consumption of the Hammer processors will be. The only thing that is for certain is that they will use a LOT less power than their Intel counterparts (which, in this particular case, are the Itaniums, which use roughly twice as much power as the hottest running Athlons). IBM's Power4 chips also pump out huge amounts of heat, so relative to these competitors chips, AMD's Opteron could be a fairly low-power solution.
The extra pins are essentially all going to be for power and ground. As these chips get more and more transistors, they're going to require more and more pins for power and ground. This, actually, could quickly start to become a major problem for chip designers, particularly when you look at the Sledgehammer, which has something like 940 pins already (note: the extra pins of the sledge vs. clawhammer are almost all I/O pins).
Ugg, I certainly HOPE that AMD doesn't fall to National Semi's hands, if it did they might botch things nearly as badly as they did with Cyrix!
Virtually every chip manufacturer in the world has already tried to make Intel-compatible chips, including National Semi on more than one occasion. Every single one of them has failed EXCEPT for AMD (even Cyrix is dead now, the chips VIA produced under the "Cyrix" name are designed by the old IDT/Centaur/Winchip design team).
The interesting part about this is that the memory controller on the Clawhammer only supports TWO memory slots when using unbuffered memory! (4 slots with buffered DIMMs).
In other words, don't plan on actually USING the third slot, it's just there for looks (sorta like the vacuum tubes:> ). Actually the third slot MIGHT work sometimes, particularly if you use some sticks of single-sided memory.
If you want lots of RAM, look for the Sledgehammers. They'll support up to 8 slots of buffered DDR memory per processor. A dual-processor Sledgehammer system should have little trouble handling 16GB of memory.
It's a wonder they haven't tried combining the two...how about a windowed hood on a Civic, or a nitrous bottle in an overclocked Celeron?
Don't joke too loudly, on a recent TV show which was showcasing a bunch of rice rockets, one of them had a plexiglass window on the hood so that you could look into the engine! No joke!
As for the nitrous, haven't quite seen that, but some of the wacky Japanese overclockers have used liquid nitrogen to super-cool their processors so that they could clock 'em higher!:>
Hmm.. still looking for a good "Type R" sticker to put on my Yugo!
Uhh, you might want to check AMD's product line-up again, or more specifically, the AMD-8131 PCI-X tunnel.
I don't know if this chip supports 64-bit PCI, though I would assume that it does on it's "legacy" 33 and 66MHz PCI bus. Of course, the main use for it will be PCI-X, which is faster than even 64-bit/66MHz PCI.
The trick though, will be that you won't be able to (easily) connect both the 8151 AGP tunnel and the 8131 PCI-X tunnel onto a Clawhammer processor, which only has a single 16-bit Hypertransport link. The Sledgehammer, with it's three 16-bit hypertransport links should have no trouble here.
You think that's bad, I still haven't been able to find a replacement for the failed fan that came INSIDE OF MY CD-RW DRIVE!
Talk about a pain in the ass, I couldn't even figure out which fan was failing for the longest time. It was making all kinds of noise, but I couldn't even see the fan. It wasn't until I tried unplugging every device in my PC one by one that I found that the CD-RW was the cause of the noise, and than I discovered this TINY little fan inside of it!
The only place I've found that sells the right sized fans is Digi-key, and they cost $30 + s/h:
On the Hammer, there's no such thing as a "northbridge".
Actually, there isn't all that much of a "northbridge" in most of today's chipsets either. Ironically enough, these days the "southbridge" is really more of a northbridge than the northbridge is.
The reason for all of this is that the "bridge" terminology is all defined with regards to the PCI bus. The northbridge is sort of where the PCI bus starts and where it has a bridge to the host controller. The southbridge is hanging off the other end of the PCI bus and provides a bridge to a different bus (ie the ISA bus).
Now though, the layout of the chipsets is quite different, and we have the PCI bus starting off of the I/O chip (ie the ICH in Intel chipsets, or the MCP in nVidia chipsets). These chips than connect using another interconnect (Intel's Accerlated Hub Architecture, Hypertransport, V-Link, etc.) to a memory controller, AGP and processor bus chip (Intel's MCH, nVidia's IGP).
With the Hammer, they stretch the separation a little bit more by moving the memory controller onto the processor itself.
Ohh, and as for the original poster, active cooling is a BAD thing IMO. Those little fans are quite possibly the least reliable part of an entire PC. If at all possible, they should be avoided from a reliability standpoint. Mind you, if you're overclocking, reliability is probably not your #1 concern. As for me though, give me a nice big passive heatsink anyday.
They aren't playing to the audiophile market at all, what they're playing to is the totally clueless consumer market.
"OHH, look, vacuum tubes! My friend has a guitar amp and says that tubes amps are the best, so this has GOT to be good!"
Vacuum tubes on a motherboard makes absolutely no sense at all. Tubes have their place in guitar amps, where the distortion they produce couples nicely with the sound of the guitar, but that's about it. At the best of times tubes really struggle to reproduce sounds as accurately as solid state amps, and when you throw them into the crowded confines of a PC case with all the EMI going around in there, they'll be piss-poor! Couple that with the fact that they're more expensive and use significantly more power, and you have an absolutely terrible solution for audio.
Uhh, I think you missed the point of the Spec benchmarks somewhere along the line!
They will not be rewritten to support anything! The whole idea of the bunchmark is to test how processors can run EXISTING CODE! And not just any old code, they test snippets of real-world code that someone wrote and actually used (ie one of the benchmarks is gzip).
The trick is, when will the compilers start to make use of AltiVec (or SSE2 for that matter). At the moment they don't do so to any great degree, and until they do, these vector units are mostly going unused since the code would have to be programmed in assembly to make use of them. And lets face it, programming in assembly just isn't practical for most large applications (despite the fact that most of the work I do involves writting assembly code:> ).
Basically what you just described above is Transmeta's Crusoe processor but able to change a bit more dynamically. A nice idea, but not overly practical. Even with the slightly simpler Crusoe design they aren't able to get even close when it comes to performance (hell, even Mac's have the Crusoe beat on the performance front! That should tell you something about just how slow they are!:> )
Ok, maybe they aren't quite the same thing yet, but the lines between the two have REALLY blurred.
Just take a look at any modern RISC processor. Chances are it has several hundred instructions, ie they sure haven't "reduced" that instruction set by any significant amount. Than if you look at any modern CISC processor, you'll find that they just decode instructions into RISC-like ops internally. End result? The difference between RISC and CISC is REAL small these days.
If you read about the design of the Power4 vs. the Athlon, you'll see that essentially ALL of the basic building blocks are the same, it's mainly just a matter of how many of those blocks there are and how they all fit together. If anyone thinks that the Power4 is so fast clock for clock vs. the Athlon is because of it's instruction set, they probably just haven't looked to see that this chip has tons of execution units, HUGE cache and a shitload of bandwidth. All things that could potentially be added to a chip like the Athlon if the economics of such would fit.
Now, this isn't to say that x86 isn't without it's flaws, but most of those flaws are rather minor and have been worked around in compilers for years. The two biggest problems are the small number of registers and the stack-based floating point units. Well, Intel's SSE2 can now mostly replace the old floating point unit for the majority of tasks (though it typically isn't used as such yet), and AMD's upcoming Hammer/Operaton will double the number of registers available.
I'm not really sure that this strategy makes any sense, or at least not anymore.
The fastest AMD chips selling today are just south of 2.0GHz. So, but your estimates, you would now buy a 1GHz chip. However, the difference in cost between a 1.0GHz chip and a 1.7GHz chips is not all that great these days, about $50. Once you factor in the cost of a motherboard and memory (which you often have to upgrade alongside your processor), this is a rather small savings for quite a large difference in performance.
I don't know if he made the whole things up, but there were some fairly major issues that should be blatently obvious to anyone watching that video and reading the explination.
First off, that whole 1 C/s temp change max makes absolutely no sense at all! If someone actually told him that then they were either completely clueless, or whoever designed the heat-protection circuitry was going out of their way to try and make it a bad design. AMD has reference specs for this on their webpage, and it's REALLY not complicated!
And than there was the P4 side of things. That P4 which apparentely stayed at 29C without it's heatsink on, because of it's thermal throttling. Think about that for a second. If your processor was only at 29C, it WOULD NOT BE THROTTLING! The thermal throttling of the P4 doesn't come into effect until somewhere around 60C, if the temp is less than 60C, the chip would run at full speed, and it would get a LOT hotter than 29C if that were the case.
Than there's simply the fact that the whole test is rather contrieved. People just don't rip heatsinks off their processors while they are running. Heatsinks don't fall off unless they were installed by a moron.
So, what we end up with is a contrieved "test" that has at least one major and very obvious flaw, not to mention a rather dubious explination for the results.
Ohh, but it DID get Tom probably a few hundred million page hits, and therefore probably several million dollars worth of advertising revenue.
First off, if your system was "blowing away" your sister's, than yes, it was mainly due to software cruft on her system. In terms of raw performance, the 450MHz G4 was about on-par with a 550MHz PII. Unfortunately for Apple, they've fallen rather quickly behind the ball since than.
Now the fastest processors that Apple sells are 1.25GHz chips, while AMD and Intel are just phasing out their 1.3GHz chips because that is their bottom end. Most of what both companies are selling are up around 2.0GHz these days.
Face it, there's a VERY simple reason why Apple NEVER releases ANY Spec scores for their processors, they're processors are SLOW! All you'll ever see from Apple are things like Photoshop "benchmarks", which are mostly invalid because Photoshop is BY FAR the easiest "benchmark" to manipulate to favor one processor over another.
Geez, does ANYONE bother to know what the hell they're talking about before posting? If so, you would KNOW that the power consumption of AMD's Athlon processors and Intel's P4 processors is damn near identical. Take a look at the datasheets! It's not like this is some sort of super-secret information known only to the Stone Cutters!
For those that are wondering, the max power consumption of an AthlonXP 2600+ is about 68W, while the max power consumption of a 2.6GHz P4 is about 65W.
Ugg.. I do WISH that people would stop reading "Tom's Hardware", or at least that they would get a clue first and realize that Tom doesn't know dick-all about what he's talking about most of the time.
His comments about heat rising more then 1C/second make NO SENSE AT ALL! It's flat-out wrong! I don't know what orafice he pulled that comment from, but it certainly had no technical backing to it. The chip uses a thermal diode. It will tell you the temperature whenever you poll it. It doesn't matter how fast or slow you poll it, it will give you the temp. You would really have to go out of your way to try to break this sort of data to get it to only be able to handle a 1C/s temp increase.
As for the heat "problem". AMD's AthlonXP chips have a maximum power consumption of roughly 50-70W. Intel's P4's have a maximum power consumption of roughly 50-70W (yes, they consume almost the exact same amount of power, check the data sheets).
For comparison, Intel's Itanium has a maximum power consumption of around 100-130W, and IBM's Power4 is also on the high-side of 100W.
A 2U case?! What? Are you NUTS?! Have you SEEN a 2U case? Those things may only be 9cm tall, but they're 48cm wide and more importantly, they're usually around least 60cm deep!
Here's a better solution, get a MiniITX case with a VIA Eden inside (ideally one with a PCI riser card so that you can put a better sound card in). Total cost would be about $200. It's small, more than fast enough for what is needed, and will run without a fan.
As for the WinXP vs. Linux debate, the main downside to using WinXP for this application is that the license costs about $200, thereby doubling the price of the box.
Slashdot Mirror
Err, IBM didn't just license Cyrix's chips, they MADE the chips for Cyrix! After a while they started keeping a few of the chips that they made and sold them under their own brand name, though I really don't think that they undercut them in price. In fact, I'm pretty certain that the IBM branded Cyrix chips were typically more expensive than the Cyrix branded ones, even though they were identical (one minor exception, I think they had slightly different core clock speed and bus speeds/clock multipliers on a few models).
What killed Cyrix had nothing to do with IBM, and everything to do with the fact that they didn't have enough money to make it as an independant company, and then they were sold to National Semiconductor in what has to have been one of the most badly botched sales in microprocessor history. Both companies ended up suffering badly as a result. NS finally gave up on the money-pit named Cyrix and essentially sold the name to VIA (the whole company was sold, but VIA than proceeded to sack pretty much the entire Cyrix team and ditched all their processor designs, than slapped the Cyrix name on the IDT Winchip processor design they had purchased at about the same time). Mind you, VIA has since dropped even the Cyrix name.
So, long story short, a lot of bad business deals and mismanagement killed Cyrix, not IBM.
If you had a K6-350 that was beaten by a P166, then you had a computer that was improperly setup, or a REALLY odd-ball setup.
Even the "really weak" floating point unit of the K6 was only about 10-20% slower than the PentiumMMX in all except the most extreme situations (and, on the other end of things, there were some floating point applications that ran faster on the K6). This is especially true by the time they got to the K6-2 350MHz chip, which had a 100MHz bus speed and a slightly improved floating point unit as compared to the original K6.
The K6 had it's flaws here and there, but for the most part it performed quite well. The floating point unit wasn't pipelined, and usually took two clock cycles per instruction, while the PentiumMMX was pipelined and therefore could sometimes pump out instructions in one clock cycle, however this didn't result in all that huge of a performance difference like some people have suggested.
Actually, the original poster was correct, at least sort of.
AMD did produce 286, 386 and 486 processors that were faster than any comperable chips that Intel ever produce (16MHz 286's if my memory serves, 40MHz 386s and 133MHz 486s). However, they did so after Intel had already released their next generation of chips.
Still, AMD did have the speed advantage at one point in time. When AMD brought out their 386DX40, the fastest chips that Intel had was their 486DX25, and AMD's 40MHz 386 was almost always faster than Intel's 25MHz 486, even though the 486 was faster clock for clock. Intel soon bumped the 486 up to 33MHz, and than it was more or less toss-up between the two, though AMD gained a lot of market share since their chips and the supporting motherboards were a LOT cheaper.
The 233MHz K6 was also, on average, marginally faster than the fastest Intel chips (the PentiumMMX 200 and PPro 200MHz) when it was released. However, that point became rather moot since it took AMD about 4 months to produce more than trivial quantities of K6's at that speed grade, and Intel brought out their PII at 233-300MHz only a few weeks after the K6 was first released.
All well and good in concept, except for that fact that there isn't a single operating system, or even a compiler in existance for Transmeta's native VLIW mode. Actually, for that matter, there isn't even any documentation available for it's VLIW instructions, so even assembly or raw byte code is out.
Face it, the Transmeta Crusoe IS an x86 chip, it just uses a really odd-ball way of getting around any Intel patents. Also, with Intel's Banias processor on the near horizon, Transmeta's days are severely numbered IMO.
The Alpha (now more of an Intel chip than anything else) is not currently being mass marketed by anyone, and it never was mass marketed.
Actually, Alpha was never really marketed much at all, which is the main reason why it never did very well, despite it's technical strengths.
The original poster was quite correct, the AMD Hammer will be the first 64-bit, general purpose CPU that is mass marketed.
Heat spreaders aren't really that great when it comes to actually making the chip run cooler, all they do is exactly what their name suggests, they spread heat.
The reason why this is a good thing is that modern processors do not generate their heat evenly, so you might have the bottom left 25% of the chip generating 50W, while the other 75% of the chip is mostly idle, only generating a few watts here or there. This is a bad thing for a variety of reasons, beyond the obvious fact that one area has to be cooled more than others. When it comes to actually dissipating the heat though, what you want is to get this heat into the furthest corners of the heatsink as quickly as possible. This is why copper inlays are a good thing for heatsinks (they spread the heat away from the hot core towards the edges of the heatsink faster than aluminium does), however even aluminium conducts heat a lot better than the P4's heat spreaders do.
This sort of thing is, presumably, less of a problem for the Athlon as compared to the P4 simply due to die sizes. The P4 is a MUCH larger die (nearly twice the size) when compared to the Athlon. So, while the distance between the hottest and coldest points in an Athlon are always going to be quite small, they can be relatively large on a P4.
As for fixing the heatsink to the motherboard instead of the socket, this is a damn good thing if you ask me. AMD originally thought that this would be how things would be done with the Athlon, and specified 4 holes around the socket to do just such a thing. You can even buy heatsinks that attach this way (my Athlon is cooled by just such a heatsink, an Alpha PAL 8045), however the vast majority of heatsink designers just flat out ignored these holes and just attached to the heatsink. Intel was a bit more forcefull, and absultely required ALL heatsinks to attached either to the motherboard or even to bolt to the case itself (the first P4s, those that came in a socket 423 package, required a special case that the heatsink bolted on to). Intel's currently design of using clips to clamp the heatsink down to the motherboard is a fairly good one IMO, though it does put an awful lot of strain on the board. AMD's plan for the Hammer/Opteron is similar, and it looks like they're going to try to avoid some of the motherboard bending/strain, however it remains to be seen just how well it will work in practice.
In any case, it remains to be seen just what the power consumption of the Hammer processors will be. The only thing that is for certain is that they will use a LOT less power than their Intel counterparts (which, in this particular case, are the Itaniums, which use roughly twice as much power as the hottest running Athlons). IBM's Power4 chips also pump out huge amounts of heat, so relative to these competitors chips, AMD's Opteron could be a fairly low-power solution.
The extra pins are essentially all going to be for power and ground. As these chips get more and more transistors, they're going to require more and more pins for power and ground. This, actually, could quickly start to become a major problem for chip designers, particularly when you look at the Sledgehammer, which has something like 940 pins already (note: the extra pins of the sledge vs. clawhammer are almost all I/O pins).
Ugg, I certainly HOPE that AMD doesn't fall to National Semi's hands, if it did they might botch things nearly as badly as they did with Cyrix!
Virtually every chip manufacturer in the world has already tried to make Intel-compatible chips, including National Semi on more than one occasion. Every single one of them has failed EXCEPT for AMD (even Cyrix is dead now, the chips VIA produced under the "Cyrix" name are designed by the old IDT/Centaur/Winchip design team).
The interesting part about this is that the memory controller on the Clawhammer only supports TWO memory slots when using unbuffered memory! (4 slots with buffered DIMMs).
:> ). Actually the third slot MIGHT work sometimes, particularly if you use some sticks of single-sided memory.
In other words, don't plan on actually USING the third slot, it's just there for looks (sorta like the vacuum tubes
If you want lots of RAM, look for the Sledgehammers. They'll support up to 8 slots of buffered DDR memory per processor. A dual-processor Sledgehammer system should have little trouble handling 16GB of memory.
It's a wonder they haven't tried combining the two...how about a windowed hood on a Civic, or a nitrous bottle in an overclocked Celeron?
Don't joke too loudly, on a recent TV show which was showcasing a bunch of rice rockets, one of them had a plexiglass window on the hood so that you could look into the engine! No joke!
As for the nitrous, haven't quite seen that, but some of the wacky Japanese overclockers have used liquid nitrogen to super-cool their processors so that they could clock 'em higher!
Hmm.. still looking for a good "Type R" sticker to put on my Yugo!
Uhh, you might want to check AMD's product line-up again, or more specifically, the AMD-8131 PCI-X tunnel.
I don't know if this chip supports 64-bit PCI, though I would assume that it does on it's "legacy" 33 and 66MHz PCI bus. Of course, the main use for it will be PCI-X, which is faster than even 64-bit/66MHz PCI.
The trick though, will be that you won't be able to (easily) connect both the 8151 AGP tunnel and the 8131 PCI-X tunnel onto a Clawhammer processor, which only has a single 16-bit Hypertransport link. The Sledgehammer, with it's three 16-bit hypertransport links should have no trouble here.
You think that's bad, I still haven't been able to find a replacement for the failed fan that came INSIDE OF MY CD-RW DRIVE!
:
Talk about a pain in the ass, I couldn't even figure out which fan was failing for the longest time. It was making all kinds of noise, but I couldn't even see the fan. It wasn't until I tried unplugging every device in my PC one by one that I found that the CD-RW was the cause of the noise, and than I discovered this TINY little fan inside of it!
The only place I've found that sells the right sized fans is Digi-key, and they cost $30 + s/h
On the Hammer, there's no such thing as a "northbridge".
Actually, there isn't all that much of a "northbridge" in most of today's chipsets either. Ironically enough, these days the "southbridge" is really more of a northbridge than the northbridge is.
The reason for all of this is that the "bridge" terminology is all defined with regards to the PCI bus. The northbridge is sort of where the PCI bus starts and where it has a bridge to the host controller. The southbridge is hanging off the other end of the PCI bus and provides a bridge to a different bus (ie the ISA bus).
Now though, the layout of the chipsets is quite different, and we have the PCI bus starting off of the I/O chip (ie the ICH in Intel chipsets, or the MCP in nVidia chipsets). These chips than connect using another interconnect (Intel's Accerlated Hub Architecture, Hypertransport, V-Link, etc.) to a memory controller, AGP and processor bus chip (Intel's MCH, nVidia's IGP).
With the Hammer, they stretch the separation a little bit more by moving the memory controller onto the processor itself.
Ohh, and as for the original poster, active cooling is a BAD thing IMO. Those little fans are quite possibly the least reliable part of an entire PC. If at all possible, they should be avoided from a reliability standpoint. Mind you, if you're overclocking, reliability is probably not your #1 concern. As for me though, give me a nice big passive heatsink anyday.
They aren't playing to the audiophile market at all, what they're playing to is the totally clueless consumer market.
"OHH, look, vacuum tubes! My friend has a guitar amp and says that tubes amps are the best, so this has GOT to be good!"
Vacuum tubes on a motherboard makes absolutely no sense at all. Tubes have their place in guitar amps, where the distortion they produce couples nicely with the sound of the guitar, but that's about it. At the best of times tubes really struggle to reproduce sounds as accurately as solid state amps, and when you throw them into the crowded confines of a PC case with all the EMI going around in there, they'll be piss-poor! Couple that with the fact that they're more expensive and use significantly more power, and you have an absolutely terrible solution for audio.
Uhh, I think you missed the point of the Spec benchmarks somewhere along the line!
:> ).
They will not be rewritten to support anything! The whole idea of the bunchmark is to test how processors can run EXISTING CODE! And not just any old code, they test snippets of real-world code that someone wrote and actually used (ie one of the benchmarks is gzip).
The trick is, when will the compilers start to make use of AltiVec (or SSE2 for that matter). At the moment they don't do so to any great degree, and until they do, these vector units are mostly going unused since the code would have to be programmed in assembly to make use of them. And lets face it, programming in assembly just isn't practical for most large applications (despite the fact that most of the work I do involves writting assembly code
Basically what you just described above is Transmeta's Crusoe processor but able to change a bit more dynamically. A nice idea, but not overly practical. Even with the slightly simpler Crusoe design they aren't able to get even close when it comes to performance (hell, even Mac's have the Crusoe beat on the performance front! That should tell you something about just how slow they are! :> )
Interesting concept, not so great in practice.
Ok, maybe they aren't quite the same thing yet, but the lines between the two have REALLY blurred.
Just take a look at any modern RISC processor. Chances are it has several hundred instructions, ie they sure haven't "reduced" that instruction set by any significant amount. Than if you look at any modern CISC processor, you'll find that they just decode instructions into RISC-like ops internally. End result? The difference between RISC and CISC is REAL small these days.
If you read about the design of the Power4 vs. the Athlon, you'll see that essentially ALL of the basic building blocks are the same, it's mainly just a matter of how many of those blocks there are and how they all fit together. If anyone thinks that the Power4 is so fast clock for clock vs. the Athlon is because of it's instruction set, they probably just haven't looked to see that this chip has tons of execution units, HUGE cache and a shitload of bandwidth. All things that could potentially be added to a chip like the Athlon if the economics of such would fit.
Now, this isn't to say that x86 isn't without it's flaws, but most of those flaws are rather minor and have been worked around in compilers for years. The two biggest problems are the small number of registers and the stack-based floating point units. Well, Intel's SSE2 can now mostly replace the old floating point unit for the majority of tasks (though it typically isn't used as such yet), and AMD's upcoming Hammer/Operaton will double the number of registers available.
I'm not really sure that this strategy makes any sense, or at least not anymore.
The fastest AMD chips selling today are just south of 2.0GHz. So, but your estimates, you would now buy a 1GHz chip. However, the difference in cost between a 1.0GHz chip and a 1.7GHz chips is not all that great these days, about $50. Once you factor in the cost of a motherboard and memory (which you often have to upgrade alongside your processor), this is a rather small savings for quite a large difference in performance.
I don't know if he made the whole things up, but there were some fairly major issues that should be blatently obvious to anyone watching that video and reading the explination.
First off, that whole 1 C/s temp change max makes absolutely no sense at all! If someone actually told him that then they were either completely clueless, or whoever designed the heat-protection circuitry was going out of their way to try and make it a bad design. AMD has reference specs for this on their webpage, and it's REALLY not complicated!
And than there was the P4 side of things. That P4 which apparentely stayed at 29C without it's heatsink on, because of it's thermal throttling. Think about that for a second. If your processor was only at 29C, it WOULD NOT BE THROTTLING! The thermal throttling of the P4 doesn't come into effect until somewhere around 60C, if the temp is less than 60C, the chip would run at full speed, and it would get a LOT hotter than 29C if that were the case.
Than there's simply the fact that the whole test is rather contrieved. People just don't rip heatsinks off their processors while they are running. Heatsinks don't fall off unless they were installed by a moron.
So, what we end up with is a contrieved "test" that has at least one major and very obvious flaw, not to mention a rather dubious explination for the results.
Ohh, but it DID get Tom probably a few hundred million page hits, and therefore probably several million dollars worth of advertising revenue.
First off, if your system was "blowing away" your sister's, than yes, it was mainly due to software cruft on her system. In terms of raw performance, the 450MHz G4 was about on-par with a 550MHz PII. Unfortunately for Apple, they've fallen rather quickly behind the ball since than.
Now the fastest processors that Apple sells are 1.25GHz chips, while AMD and Intel are just phasing out their 1.3GHz chips because that is their bottom end. Most of what both companies are selling are up around 2.0GHz these days.
Face it, there's a VERY simple reason why Apple NEVER releases ANY Spec scores for their processors, they're processors are SLOW! All you'll ever see from Apple are things like Photoshop "benchmarks", which are mostly invalid because Photoshop is BY FAR the easiest "benchmark" to manipulate to favor one processor over another.
Geez, does ANYONE bother to know what the hell they're talking about before posting? If so, you would KNOW that the power consumption of AMD's Athlon processors and Intel's P4 processors is damn near identical. Take a look at the datasheets! It's not like this is some sort of super-secret information known only to the Stone Cutters!
For those that are wondering, the max power consumption of an AthlonXP 2600+ is about 68W, while the max power consumption of a 2.6GHz P4 is about 65W.
I know that people hate facts, but here they are:
Power consumption:
AMD AthlonXP 2600+ : 68.3W Max, 62.0W typical
Intel Xeon 2.4GHz: 65W TDP (*)
*TDP = Thermal Design Power, a kind of ambigious measure of power that is slightly less then the maximum power the chip can use.
Ugg.. I do WISH that people would stop reading "Tom's Hardware", or at least that they would get a clue first and realize that Tom doesn't know dick-all about what he's talking about most of the time.
His comments about heat rising more then 1C/second make NO SENSE AT ALL! It's flat-out wrong! I don't know what orafice he pulled that comment from, but it certainly had no technical backing to it. The chip uses a thermal diode. It will tell you the temperature whenever you poll it. It doesn't matter how fast or slow you poll it, it will give you the temp. You would really have to go out of your way to try to break this sort of data to get it to only be able to handle a 1C/s temp increase.
As for the heat "problem". AMD's AthlonXP chips have a maximum power consumption of roughly 50-70W. Intel's P4's have a maximum power consumption of roughly 50-70W (yes, they consume almost the exact same amount of power, check the data sheets).
For comparison, Intel's Itanium has a maximum power consumption of around 100-130W, and IBM's Power4 is also on the high-side of 100W.
A 2U case?! What? Are you NUTS?! Have you SEEN a 2U case? Those things may only be 9cm tall, but they're 48cm wide and more importantly, they're usually around least 60cm deep!
Here's a better solution, get a MiniITX case with a VIA Eden inside (ideally one with a PCI riser card so that you can put a better sound card in). Total cost would be about $200. It's small, more than fast enough for what is needed, and will run without a fan.
As for the WinXP vs. Linux debate, the main downside to using WinXP for this application is that the license costs about $200, thereby doubling the price of the box.
Damn, where do I apply to work for this company? I want to drink beer at work too! :>