Prescott was faster at some benchamarks, and prescott is cheaper to manufacture, presscot is currently PACKAGING limited in clock speed. [Those damn little pins cannot handle enough current to feed the beast, especially with REDUCED voltage.] The performance, as overall the 3.4Ghz prescott should be slightly better buy than 3.4Ghz northwood as it scales a LOT better in clock speed. Basicly if cannot move enough power to the die you cannot clock it high, so thats the new packaging that will unleash the power of prescott... Also its versio 1.0... The L1 cache numbers just look TERRIBLE, worser latency than 64Kb cache would of been, they probably fix that and performance jumps up again, or they really designed it with extreme clock speed, and perhaps a SMALL part just failed to get the high clock speed. The long pipeline isn't that bad because they improved branch predictor. Something tells me they made it run VERY high clock speeds and scale performance well with clock speed. BUT there was a little problem, and they couldn't get the high clock speed from day 1.
PS. I'll personally think getting A64 is better than these just price/performance point of view, intel is market leader and prices its products higher than equally performing underdogs products.
'2.13.5 sucks soo badly mostly since linus got addicted to heroin? we're bout to start a fork back at 2.9.3 where it was still descent' and we can do this! do you think we're ever going to be able to fork windows 95?
Hey 2.13.5 sucks, but why to hell you wan't to go 2.9.3, both of them are unstable versions, fork the stable release not just some beta crap either 2.8.6 or 2.10.5 .
Firstly they are making it enough volumes to attract others to build applications that need it, but not enough to generate any real money...But have plans for mass manufacturing plant with 12 to 24 month period from now, but they are seeking funding to build it, which means they have to show SOMETHING, with future potential. They are doing active research, meanwhile which means that they can make it bigger, last longer,cheaper, and have better resolution, they just don't know yet which of those are going to improve and how much in the period before mass production. The first generation product is probably usably mainly as prototype design and perhaps some rare cases, where cost is not such a problem...
They have 27 000 people in the mobile phones compared to only 15000 people in the networks. The and mobile phone unit is 80% of nokia sales while networking is only 20%... And with 38% of workforce in R&D as overall means that they do probably have some serious R&D in handset section too . Especially with those sales numbers... [I got em from finnish site so no link]. Actual manufacturing is outsourced ofcourse... But still nokia probably designs the most of the handsets itself, but I know they use subcontractors heavily.
Currently design are done by small cell phones design house in Europe and Korea and manufecture by Taiwanese and Chinese OEM.
So you mean that thousands of electrical engineer nokia has in its payroll in my country do nothing?
Besides it has thousands of software engineers too...
First things first, nokia do design its phones, 2ndly it has subcontractors for certain parts of a design, not complete phone... And OEM manufacturing is standard in electronics business...
I know these things because I'm studying electrical engineering in a country where Nokia and its subcontractors practicly dominates the working positions for engineers...[Finland right next to Nokia headquarters.]
Nokia is practicly dominating the market, and other top players too go for symbian... This results that there will be large global pool for symbian applications soon... Another point is that number 1 mobile phone manufacturer (NOKIA) has stake in symbian so, they won't give up on it for linux. So atleast europe, asia and middle-east will go for symbian, instead of something else...
>>Why go through all the trouble to make it 64-bit anyway? Other than sex appeal, what other reasons are there for 64-bit?
Alrighty, if someone can show me some Hot Babes who think running 64-bits gives me more sex appeal then I'll run out and buy a 64-bit system right now. I'll take a half dozzen. Hell, make them 128-bit and don't bother me with benchmarks!
Well you assumed it was hot chicks that was attracted by such sex appeal. I think its more manly thing in this matter;)
Firstly I wan't to debunk some myths on branch missprediction stalls. Firstly from pentium pro all the stalls in the pipeline combined was more than double the CPI of the computer. And thats result of OO execution, when you have large OO window the any single type of stall means less to you. Your execution engine probably has some instructions BEFORE the branch that where stalled before the branch was executed, so the stall is not mathematically equal. 2nd point is that larger OoO window means a LOT more work to deal with, so P4 requires more work to do with large OoO window, and that large OoO window is BIG part of long pipeline, but the benefits of large OoO are more related to memory latency, it gives memory subsystem more known addresses to handle in the memory pipeline which means better scalability, when increasing core execution capabilities, [either width or dept=clock it doesn't matter]. Now P4 shines in better dealing with memory latencies, and thats more important than just a branch missprediction. Also Prescot has superior branch predictor so reducing misspredicted branches negates the missprediction penalty. L1&L2 cache are doubled which means that integer performance of prescott is increased by reducing the critical code paths from going to L2 cache. And increased L2 cache reduces the amount of time spend on waiting the memory, so I'd say prescott should do just fine in IPC side of things also. In overall P4's systems aren't as good as integrated memory controller, but it just helps to tolerate the latency a lot better than the PIII or athlon design.And in overall when most of time is spend on waiting memory references the narrow and deep approach seems slightly better. [Narrow reduces the costs of increased reordering capabilities, don't put G5 reference here they patented clever trick and are risc for christ sake. So they are wide and deep for improved reordering.] . The thing I'd wan't from AMD would be more or less, one additional pipeline stages for better reordering capabilities.
Huh? Before Peter Russia had no ports with any access to Atlantic Ocean at all, so taking the northwest and founding St.Petersburg provided such an access through Baltic Sea, for the first time in many centuries of being for all practical purposes a landlocked country. One may question if such a goal would be "legal" in modern times, however in the 17th century a country with no access to the sea could just as well be located in the middle of a desert, and St. Petersburg was vital to the development of Russia. Sweden, Holland and Norway were at that point way too strong to even think attacking them, and Swedes went as far south as Ukraine (then part of the Russian Empire) when they had a war with Russia. But Russian expansion to the northwest ended quickly ather the goal of getting and securing the sea access was achieved. What would Russians gain by attacking anything in Scandinavia, I have no idea, and I guess, neither did Peter the Great (and all Czars after him) nor Stalin (and all USSR leaders after him). So finnish history books lie, as well as books in our museums? The advantage was a suport ports to atlantic sea, in Peters time, the denmarks straights could be blocked easily. <I> Yeah, right. Every piece of every country gets its share of nationalists and separatists when any kind of war hits it, and there is an opportunity to snatch a piece of land while the big guys are too busy fighting. This does not change the fact that both countries were a part of Russian Empire before USSR was formed, and the full Czar's title for many centuries read "...of Great, Small and White Russia..." (what meant Russia, Ukraine and Belarus). Civil war of 1918-22 was a great opportunity for local rebels and gangs to fight for "independence" of their stomping grounds under nationalistic slogans, however their claim to power was orders of magnitude less than one of Communists. When this kind of nationalists succeed, we see shining examples of democracy in action such as Serbia, Bosnia, Croatia, Chechnya, Georgia (one that is not a US state), 70% of Africa, and other world-famous shitholes, filled with either violent gangs, or hopelessly corrupt oppressive regimes depending on the local traditions and amounts of resources available. And the worst happens when the real aggressor (such as Nazi or US) comes -- nationalists suddenly become humble and obedient servants of the new rulers, fighting against their own people, whose "independence" was their proclaimed goal. </I> Well a country so well presecuting its minorities is not to be reckoned with.(to put in to perspective russian speakers where minority even in european side of current russia y1k, they expanded at relative speed and killed or forcefully converted most of the people to russian. For instance Finnish was given certain rights in Wien congress, when it was given to russia. In 1902-1905 they started persecution of Fins in finnish like banning Finnish education and forcing Russian education, prosecuting finnish news papers... During war time same again. In poland 1800's the rebellion was major, it was majority of people there who was in rebellion same was ukraine in 1800's. Think living under russian rule where your own language is under pressure and and people are forced to russian language. Its the prosecution tha creates the rebellions it was never a small group, since small groups wouldn't have any chance against military. It was LARGE part of non-russian population which was in most of the rebellion provinces. I think the more proper way of calling it would of been slave revolt since thats what non russian where. [Unless tzar wanted to give example to other european nations how good he was... (Finland was in 1800's the tzars show area to other nations while rest of russia was in worser for non russian natives...) I mean that in 1918 or somewhere around that (I was terrible at dates, dates where not interesting but ideas who things happened.) Germany and russian signed peace deal with gave indence to certain russias aread who where
I don't know about trotsky doctrine I just know they build roads to strategic places in order to get to finnish border and THAT was build long before, they prepared for major offensive support structure in karelia long before WWII... The baltic states was long part of russians defensive plans, peter the great had its eyes on norther norway to get good atlantic harbors, and THAT strategy was part OF stalins doctrine too. Belorussia and Ukraine was NON willingly joined to USSR it was just because red army was there, when USSR joined, and Ukraine was violently resisting the becoming part of USSR, actually they where in rebellion against russia couple of times BEFORE! And this I talk about ENDING of WWI when USSR was formed, at that point they where expansionist. Leaving baltic states in WWI when they separated from russia. They got them back by threats in WWII. BTW: I know this thing because its as important part of our history as declaration of independence and independence war is for those Americans. [JollyFinn] Finn= A Finnish person;)
Actually the USSR was preparing for expansion before WWII. They considered their responcibility to advance the communism elsewhere. And the war with Finland, and crabbing of baltic states was already planned ahead. It was just implementing the 19th century plans for defence of Leningrad... And they already made preparation years before WWII for attacking finland. Afghanistan was planned assault. Europe after WWII was out of question due the nuke problem. China had too much population to deal with, and middle east would of gotten them full scale war against Nato. So what they could of done? I mean they didn't wan't to get a full scale war against nuclear states and wanted to expand anyway. Their hands where pretty much tied up. Their navy was not as capable for non nuclear warfare as NATO so getting a naval assault on small country was not an option. It was simple. Before WWII they didn't have industry and military prepare for war. And after US got nukes, and there was strong enough allience to stop em with conventional weapons at non european fronts. Europe is a No, no for soviet expansion because of nato... Now what comes next... Middle east? Well thats US oil, and they properly estimated the result of getting a war in there. Next central asia, well they tried in afganistan, and last is china/mongolia, both communist regimes and war against mongolia would result war against china, with US supplying chinese infantry with weapons... They expanded during WWII and tried to hold that as much as possible and tryid to expand also but failed. Basicly there was EXPANSIONIST individuals in power for some time. Like stalin and lenin... (Yes Lenin was expansionist he just though that the country was not ready for it and was correct.) Finlands,baltic states, belorussians and Ukranian s separation was part of a peace deal between russia and germany , and the war that formed USSR got two of those states back. One because it was base operations of the other fraction and other just because... The smaller states was left intact, just because the internal image would of hurt if they wouldn't shown that they kept their words, until the controlling system was build up.
> Could it be that American programmers spoke the same language as the managers so were better able to communicate.
What do nerds learned marketing speach and management hype, thats quite a statement to make. I'm not convinced that MBA:s and Msc:s speak same language... Think again...
Microsoft simply defines the overall architecture and some parts while gets real world players to implement. This was the reason they got PPC and IBM for XBOX2, as I mentioned then. The 970 with several customizations put in place some custom asic functions attached to it and SEPARATE GFX chip would be the thing. The GFX should be separated for cooling issues, we don't wan't these two hot cores next to each other. The decoding signatures for software may be integrated in hardware and THAT part MS would probably be interested in mostly. So you wouldn't be able to run non signatured binaries on it. So as MS controls the signaturing XBOX2 is unreachable for non certified software. But I think still its some of its partners that do the hardwork here. MS will simply just gets more on the IP rights of the work. And its more active simply means that they will define what they want others to make them, while NOT taking a standard parts from them. So don't expect running your favourite OS on XBOX 2, and for memory bandwith issue. Well they could go a specialized memory that is made for highbandwith embedded enviroment like XBOX. 3.2Ghz memory interface running 64 bits at a time could be the thing, for XBOX 2 that separates it from non custom chips by far. IBM is a large ASIC company that integrates CPU:s in those asics so this is not anyway new to them.
Blah. They are probably far more than 6 months away from the business. So that means that atleast ONE process shrink that improves clock speed, and changing the endianess of a MMU isn't hard at all. MS could simply ask IBM to make em little endian, and thats it. But I think they integrate memory controller on chip also. The G5 could be the core for XBox2 but espect it heavily modified for MS needs. What 4 channel 3.2ghz XRDRAM controller ondie... OR MCM with memory chip and 8 channel memory interface...
And new vector extensions spefically for XBOX2, doing 16FMAC:s/cycle, and... All these things are possible, and wouldn't be overly expensive to do. 16(32bit)FMACS ~ 1/2M transistors... You know why special purpose chips kick ass for CPU:s and IBM probably offered them a special purpose chip for XBOX2. What you don't WANT memory as upgradable modules well that simply means they can do solutions that multiply their memory bandwith compared to PC:s. What about the costs.... Well mass processed wafers costs 3000$ for And they have (150mm)*PI area... Expect yields relaticely close to 100% for 100mm chip. Lets round down and take that rounding as the yild loss lets say they get 500 peaces of 100mm die:s per wafer thats 6$ per chip, the packaging, testing and designer time is more expensive than pure wafer area. At 0.9u they could put 2MB of L2 cache and TWO G5 cores there from cost point of view. On the otherhand the power consumption will keep them from putting the 2nd core. So they probably put few special function accelerations on as extra instructions or separate cores, change endianess, add ondie memory controllers and put a Lot of cache, as cache consumes order of magnitude less power than logic. And special purpose logic is order of magnitude more energy efficient than doing same things in general purpose chip. IBM does a lot custom chips while Intel and AMD is more of producers of standard chips in mass quantities. To put it in perspective, they get few customizations and 2Ghz core there they probably beat G4 in x86 emulation by far. 2Ghz G5 with little endian mode by default 2MB of L2 cache with improved bandwith, and 24GB/s of memory bandwith available is not out of question for XBOX2. Think that trying to emulate 733 celeron, 3 times clockspeed 8 instructions/cycle/vs 3... a Lot more cache and a lots more memory bandwith , the difference should be enough to emulate it in real time. Why they would get 24GB/s memory bandwith, its special purpose memorychips on board just like Nvidia and ATI and not just upgradable modules that cause interfaces pain in the ass signaling problems.
IBM microelectronics make custom asics with PPC cores in them, and IBM:s chip designing is in higher level stuff than AMD so they can modify that cheaper at expense of clock speed that they get... But now at 0.9u the PPC970 is supposed to be quite tiny so what ELSE they will put in there besides the CPU core and cache? Instead of using altivec they might go something more excess like putting 16 FMACS. Which would give microsoft both superiour numbers and performance but also guarantee that other chips wouldn't be compatible with it, as they would have instructions that no one else has, and in other way their developement package might be really only way to port software for it, and the customizations might even make reverse engineering the thing without full developement package from microsoft impossible. They could offer packaging with low latency mainmemory in the package, and something like 4-8 channels to the memory chip, inside the package. And only put outside interface to graphic chip outside the package and put all the other supporting logic in the same chip with CPU. Hey IBM has LOTS of options and modifications and stuff that they could have offered for microsoft besides price point. IBM could have made point hey we offer you 4 times as much memory bandwith and 4 times as many flops as our competitors in same price if you take the reduction of other chips in the system in account. And AMD and INTEL in their highly tuned hand optimized design methologies where not able to offer something even resembling the beast that IBM could customize for microsoft, at reasonable price. IBM makes great business selling G3:s with lots of custom stuff attached to it on single chip. They might even maker HARDWARE decryption on the processor chip for instructions stream, that could mean a LOT harder modifications for it than for original xbox.
G3 is dead for desktop, and in many other places. Enchanched G3 chips are aimed at in EMBEDDED market, thats right in printers in PDA:s and other such mediums where low power comsumption low price and sometimes ability to put custom logic on same chip with processor core may help. Thats where most of PPC:s go for not apple nor any IBM workstations/servers but for other products with different companies.
Original alphas where cheaper to manufacture than original pentiums, even with similar processes! Alpha:s cost about 200$ to make while pentium:s where over 500$ a piece. They just didn't make cheap entrylevel machines, they where supposed to replace other EXPENSIVE peace of equipment like mini computers and workstations,and the management didn't really try to compete against PC:s so they died on onslaught of the killer micros.
Microsoft has TWO products which make them money everything else just helps them control the market. One is windows and other is office. Rest means nothing in terms of cash flow, now with staroffice in linux and more people turning to linux would M$ even think turning to linux and practicly loosing their profits... They stick to windows longer than possible, if they start loosing money on windows then MS is in deeper trouble than you could even imagine although the huge pile of cash they already accumulated could help them create profitable business before they burn it...
Slashdot Mirror
And because programmers are forced to do it anyway... For programmers open source would be medical term.
PS. I'll personally think getting A64 is better than these just price/performance point of view, intel is market leader and prices its products higher than equally performing underdogs products.
Thats special purpose designed screwdriver that handles a situation where normal screwdriver just won't cut it.
Hey 2.13.5 sucks, but why to hell you wan't to go 2.9.3, both of them are unstable versions, fork the stable release not just some beta crap either 2.8.6 or 2.10.5 .
Firstly they are making it enough volumes to attract others to build applications that need it, but not enough to generate any real money...But have plans for mass manufacturing plant with 12 to 24 month period from now, but they are seeking funding to build it, which means they have to show SOMETHING, with future potential. They are doing active research, meanwhile which means that they can make it bigger, last longer ,cheaper, and have better resolution, they just don't know yet which of those are going to improve and how much in the period before mass production. The first generation product is probably usably mainly as prototype design and perhaps some rare cases, where cost is not such a problem...
They have 27 000 people in the mobile phones compared to only 15000 people in the networks.
The and mobile phone unit is 80% of nokia sales while networking is only 20%...
And with 38% of workforce in R&D as overall means that they do probably have some serious R&D in handset section too . Especially with those sales numbers... [I got em from finnish site so no link].
Actual manufacturing is outsourced ofcourse... But still nokia probably designs the most of the handsets itself, but I know they use subcontractors heavily.
So you mean that thousands of electrical engineer nokia has in its payroll in my country do nothing? Besides it has thousands of software engineers too... First things first, nokia do design its phones, 2ndly it has subcontractors for certain parts of a design, not complete phone... And OEM manufacturing is standard in electronics business... I know these things because I'm studying electrical engineering in a country where Nokia and its subcontractors practicly dominates the working positions for engineers...[Finland right next to Nokia headquarters.]
Nokia is practicly dominating the market, and other top players too go for symbian... This results that there will be large global pool for symbian applications soon... Another point is that number 1 mobile phone manufacturer (NOKIA) has stake in symbian so, they won't give up on it for linux. So atleast europe, asia and middle-east will go for symbian, instead of something else...
Alrighty, if someone can show me some Hot Babes who think running 64-bits gives me more sex appeal then I'll run out and buy a 64-bit system right now. I'll take a half dozzen. Hell, make them 128-bit and don't bother me with benchmarks!
Well you assumed it was hot chicks that was attracted by such sex appeal. I think its more manly thing in this matter ;)
Firstly I wan't to debunk some myths on branch missprediction stalls. Firstly from pentium pro all the stalls in the pipeline combined was more than double the CPI of the computer. And thats result of OO execution, when you have large OO window the any single type of stall means less to you. Your execution engine probably has some instructions BEFORE the branch that where stalled before the branch was executed, so the stall is not mathematically equal. 2nd point is that larger OoO window means a LOT more work to deal with, so P4 requires more work to do with large OoO window, and that large OoO window is BIG part of long pipeline, but the benefits of large OoO are more related to memory latency, it gives memory subsystem more known addresses to handle in the memory pipeline which means better scalability, when increasing core execution capabilities, [either width or dept=clock it doesn't matter].
Now P4 shines in better dealing with memory latencies, and thats more important than just a branch missprediction. Also Prescot has superior branch predictor so reducing misspredicted branches negates the missprediction penalty. L1&L2 cache are doubled which means that integer performance of prescott is increased by reducing the critical code paths from going to L2 cache. And increased L2 cache reduces the amount of time spend on waiting the memory, so I'd say prescott should do just fine in IPC side of things also.
In overall P4's systems aren't as good as integrated memory controller, but it just helps to tolerate the latency a lot better than the PIII or athlon design.And in overall when most of time is spend on waiting memory references the narrow and deep approach seems slightly better. [Narrow reduces the costs of increased reordering capabilities, don't put G5 reference here they patented clever trick and are risc for christ sake. So they are wide and deep for improved reordering.]
.
The thing I'd wan't from AMD would be more or less, one additional pipeline stages for better reordering capabilities.
He just don't do stuff that matters.
He does the important stuff...
Huh? Before Peter Russia had no ports with any access to Atlantic Ocean at all, so taking the northwest and founding St.Petersburg provided such an access through Baltic Sea, for the first time in many centuries of being for all practical purposes a landlocked country. One may question if such a goal would be "legal" in modern times, however in the 17th century a country with no access to the sea could just as well be located in the middle of a desert, and St. Petersburg was vital to the development of Russia. Sweden, Holland and Norway were at that point way too strong to even think attacking them, and Swedes went as far south as Ukraine (then part of the Russian Empire) when they had a war with Russia. But Russian expansion to the northwest ended quickly ather the goal of getting and securing the sea access was achieved. What would Russians gain by attacking anything in Scandinavia, I have no idea, and I guess, neither did Peter the Great (and all Czars after him) nor Stalin (and all USSR leaders after him).
So finnish history books lie, as well as books in our museums? The advantage was a suport ports to atlantic sea, in Peters time, the denmarks straights could be blocked easily.
<I> Yeah, right. Every piece of every country gets its share of nationalists and separatists when any kind of war hits it, and there is an opportunity to snatch a piece of land while the big guys are too busy fighting. This does not change the fact that both countries were a part of Russian Empire before USSR was formed, and the full Czar's title for many centuries read "...of Great, Small and White Russia..." (what meant Russia, Ukraine and Belarus). Civil war of 1918-22 was a great opportunity for local rebels and gangs to fight for "independence" of their stomping grounds under nationalistic slogans, however their claim to power was orders of magnitude less than one of Communists. When this kind of nationalists succeed, we see shining examples of democracy in action such as Serbia, Bosnia, Croatia, Chechnya, Georgia (one that is not a US state), 70% of Africa, and other world-famous shitholes, filled with either violent gangs, or hopelessly corrupt oppressive regimes depending on the local traditions and amounts of resources available. And the worst happens when the real aggressor (such as Nazi or US) comes -- nationalists suddenly become humble and obedient servants of the new rulers, fighting against their own people, whose "independence" was their proclaimed goal. </I>
Well a country so well presecuting its minorities is not to be reckoned with.(to put in to perspective russian speakers where minority even in european side of current russia y1k, they expanded at relative speed and killed or forcefully converted most of the people to russian. For instance Finnish was given certain rights in Wien congress, when it was given to russia. In 1902-1905 they started persecution of Fins in finnish like banning Finnish education and forcing Russian education, prosecuting finnish news papers... During war time same again. In poland 1800's the rebellion was major, it was majority of people there who was in rebellion same was ukraine in 1800's. Think living under russian rule where your own language is under pressure and and people are forced to russian language. Its the prosecution tha creates the rebellions it was never a small group, since small groups wouldn't have any chance against military. It was LARGE part of non-russian population which was in most of the rebellion provinces. I think the more proper way of calling it would of been slave revolt since thats what non russian where. [Unless tzar wanted to give example to other european nations how good he was... (Finland was in 1800's the tzars show area to other nations while rest of russia was in worser for non russian natives...)
I mean that in 1918 or somewhere around that (I was terrible at dates, dates where not interesting but ideas who things happened.) Germany and russian signed peace deal with gave indence to certain russias aread who where
I don't know about trotsky doctrine I just know they build roads to strategic places in order to get to finnish border and THAT was build long before, they prepared for major offensive support structure in karelia long before WWII... The baltic states was long part of russians defensive plans, peter the great had its eyes on norther norway to get good atlantic harbors, and THAT strategy was part OF stalins doctrine too. Belorussia and Ukraine was NON willingly joined to USSR it was just because red army was there, when USSR joined, and Ukraine was violently resisting the becoming part of USSR, actually they where in rebellion against russia couple of times BEFORE! . ;)
And this I talk about ENDING of WWI when USSR was formed, at that point they where expansionist. Leaving baltic states in WWI when they separated from russia. They got them back by threats in WWII
BTW: I know this thing because its as important part of our history as declaration of independence and independence war is for those Americans. [JollyFinn] Finn= A Finnish person
Actually the USSR was preparing for expansion before WWII. They considered their responcibility to advance the communism elsewhere. And the war with Finland, and crabbing of baltic states was already planned ahead. It was just implementing the 19th century plans for defence of Leningrad... And they already made preparation years before WWII for attacking finland. Afghanistan was planned assault.
Europe after WWII was out of question due the nuke problem. China had too much population to deal with,
and middle east would of gotten them full scale war against Nato. So what they could of done?
I mean they didn't wan't to get a full scale war against nuclear states and wanted to expand anyway. Their hands where pretty much tied up. Their navy was not as capable for non nuclear warfare as NATO so getting a naval assault on small country was not an option. It was simple.
Before WWII they didn't have industry and military prepare for war. And after US got nukes, and there was strong enough allience to stop em with conventional weapons at non european fronts. Europe is a No, no for soviet expansion because of nato... Now what comes next... Middle east? Well thats US oil, and they properly estimated the result of getting a war in there. Next central asia, well they tried in afganistan, and last is china/mongolia, both communist regimes and war against mongolia would result war against china, with US supplying chinese infantry with weapons...
They expanded during WWII and tried to hold that as much as possible and tryid to expand also but failed. Basicly there was EXPANSIONIST individuals in power for some time. Like stalin and lenin...
(Yes Lenin was expansionist he just though that the country was not ready for it and was correct.)
Finlands,baltic states, belorussians and Ukranian s separation was part of a peace deal between russia and germany , and the war that formed USSR got two of those states back. One because it was base operations of the other fraction and other just because... The smaller states was left intact, just because the internal image would of hurt if they wouldn't shown that they kept their words, until the controlling system was build up.
> Could it be that American programmers spoke the same language as the managers so were better able to communicate.
What do nerds learned marketing speach and management hype, thats quite a statement to make. I'm not convinced that MBA:s and Msc:s speak same language... Think again...
No they are not version numbers they are relative bug counts for each product. And for two of the latest the 32bits was not enough for counting.
>> If a conventional train derails at 60mph you're screwed. Uhh, I didn't know women where attracted to train accidents...
Microsoft simply defines the overall architecture and some parts while gets real world players to implement. This was the reason they got PPC and IBM for XBOX2, as I mentioned then. The 970 with several customizations put in place some custom asic functions attached to it and SEPARATE GFX chip would be the thing. The GFX should be separated for cooling issues, we don't wan't these two hot cores next to each other. The decoding signatures for software may be integrated in hardware and THAT part MS would probably be interested in mostly. So you wouldn't be able to run non signatured binaries on it. So as MS controls the signaturing XBOX2 is unreachable for non certified software. But I think still its some of its partners that do the hardwork here. MS will simply just gets more on the IP rights of the work. And its more active simply means that they will define what they want others to make them, while NOT taking a standard parts from them. So don't expect running your favourite OS on XBOX 2, and for memory bandwith issue. Well they could go a specialized memory that is made for highbandwith embedded enviroment like XBOX. 3.2Ghz memory interface running 64 bits at a time could be the thing, for XBOX 2 that separates it from non custom chips by far. IBM is a large ASIC company that integrates CPU:s in those asics so this is not anyway new to them.
Blah. They are probably far more than 6 months away from the business. So that means that atleast ONE process shrink that improves clock speed, and changing the endianess of a MMU isn't hard at all. MS could simply ask IBM to make em little endian, and thats it. But I think they integrate memory controller on chip also. The G5 could be the core for XBox2 but espect it heavily modified for MS needs. What 4 channel 3.2ghz XRDRAM controller ondie ... OR MCM with memory chip and 8 channel memory interface... /cycle, and ... All these things are possible, and wouldn't be overly expensive to do. 16(32bit)FMACS ~ 1/2M transistors... You know why special purpose chips kick ass for CPU:s and IBM probably offered them a special purpose chip for XBOX2. What you don't WANT memory as upgradable modules well that simply means they can do solutions that multiply their memory bandwith compared to PC:s. /vs 3 ... a Lot more cache and a lots more memory bandwith , the difference should be enough to emulate it in real time.
And new vector extensions spefically for XBOX2, doing 16FMAC:s
What about the costs....
Well mass processed wafers costs 3000$ for And they have (150mm)*PI area... Expect yields relaticely close to 100% for 100mm chip. Lets round down and take that rounding as the yild loss lets say they get 500 peaces of 100mm die:s per wafer thats 6$ per chip, the packaging, testing and designer time is more expensive than pure wafer area. At 0.9u they could put 2MB of L2 cache and TWO G5 cores there from cost point of view. On the otherhand the power consumption will keep them from putting the 2nd core. So they probably put few special function accelerations on as extra instructions or separate cores, change endianess, add ondie memory controllers and put a Lot of cache, as cache consumes order of magnitude less power than logic. And special purpose logic is order of magnitude more energy efficient than doing same things in general purpose chip.
IBM does a lot custom chips while Intel and AMD is more of producers of standard chips in mass quantities.
To put it in perspective, they get few customizations and 2Ghz core there they probably beat G4 in x86 emulation by far.
2Ghz G5 with little endian mode by default 2MB of L2 cache with improved bandwith, and 24GB/s of memory bandwith available is not out of question for XBOX2. Think that trying to emulate 733 celeron, 3 times clockspeed 8 instructions/cycle
Why they would get 24GB/s memory bandwith, its special purpose memorychips on board just like Nvidia and ATI and not just upgradable modules that cause interfaces pain in the ass signaling problems.
All you guys who wan't blender become userfriendly think again do you really want Erwin.
IBM microelectronics make custom asics with PPC cores in them, and IBM:s chip designing is in higher level stuff than AMD so they can modify that cheaper at expense of clock speed that they get... But now at 0.9u the PPC970 is supposed to be quite tiny so what ELSE they will put in there besides the CPU core and cache? Instead of using altivec they might go something more excess like putting 16 FMACS. Which would give microsoft both superiour numbers and performance but also guarantee that other chips wouldn't be compatible with it, as they would have instructions that no one else has, and in other way their developement package might be really only way to port software for it, and the customizations might even make reverse engineering the thing without full developement package from microsoft impossible. They could offer packaging with low latency mainmemory in the package, and something like 4-8 channels to the memory chip, inside the package. And only put outside interface to graphic chip outside the package and put all the other supporting logic in the same chip with CPU. Hey IBM has LOTS of options and modifications and stuff that they could have offered for microsoft besides price point. IBM could have made point hey we offer you 4 times as much memory bandwith and 4 times as many flops as our competitors in same price if you take the reduction of other chips in the system in account. And AMD and INTEL in their highly tuned hand optimized design methologies where not able to offer something even resembling the beast that IBM could customize for microsoft, at reasonable price. IBM makes great business selling G3:s with lots of custom stuff attached to it on single chip. They might even maker HARDWARE decryption on the processor chip for instructions stream, that could mean a LOT harder modifications for it than for original xbox.
G3 is dead for desktop, and in many other places. Enchanched G3 chips are aimed at in EMBEDDED market, thats right in printers in PDA:s and other such mediums where low power comsumption low price and sometimes ability to put custom logic on same chip with processor core may help. Thats where most of PPC:s go for not apple nor any IBM workstations/servers but for other products with different companies.
>Have we even found the origin of AIDS yet? Are we 100% certian that it was not engineered?
Thats why it infected GAY population first in USA?
Original alphas where cheaper to manufacture than original pentiums, even with similar processes!
Alpha:s cost about 200$ to make while pentium:s where over 500$ a piece. They just didn't make cheap entrylevel machines, they where supposed to replace other EXPENSIVE peace of equipment like mini computers and workstations,and the management didn't really try to compete against PC:s so they died on onslaught of the killer micros.
Microsoft has TWO products which make them money everything else just helps them control the market.
One is windows and other is office. Rest means nothing in terms of cash flow, now with staroffice in linux and more people turning to linux would M$ even think turning to linux and practicly loosing their profits... They stick to windows longer than possible, if they start loosing money on windows then MS is in deeper trouble than you could even imagine although the huge pile of cash they already accumulated could help them create profitable business before they burn it...