OneMike
Mar 26, 06:27 PM
No Rosetta, no sale for me. Not ready to move on.
I'm glad rosetta is going away. Maybe the dev will finally update the app.
I'm glad rosetta is going away. Maybe the dev will finally update the app.
swingerofbirch
Aug 26, 07:40 PM
I'm sure the GPU will also be bumped, at the very least. The MBP will probably also see some things that the MB has like a user-removable hard drive and magnetic latch. The CPU and GPU alone make it worth getting the new one, IMO.
Also, I'll say it one last time (yea right) - the imac should not and will not get a mobile processor. It only got Yonah because there was no alternative. It had a real desktop processor when one was available on the PPC side (G5), and it will have a real desktop processor now that one is available on the intel side (Conroe). Leave merom for what it was meant for - laptops.
I agree with you wholeheartedly.
But, I guess they COULD have put a pentium d in them...didnt they have dual cores?
Also, I'll say it one last time (yea right) - the imac should not and will not get a mobile processor. It only got Yonah because there was no alternative. It had a real desktop processor when one was available on the PPC side (G5), and it will have a real desktop processor now that one is available on the intel side (Conroe). Leave merom for what it was meant for - laptops.
I agree with you wholeheartedly.
But, I guess they COULD have put a pentium d in them...didnt they have dual cores?
brianus
Sep 13, 10:05 AM
probably due to latency involved in distributing the load across the two processors. that's the same problem a single Clovertown would have. Only true quads wouldn't suffer from these problems (earliest seems to be Harpertown in don't know if there are any non-MCM Xeons scheduled before then)
What about Tigerton (2007)? Isn't that a "true" quad?
What about Tigerton (2007)? Isn't that a "true" quad?
Multimedia
Jul 27, 11:48 PM
Ahh so many announcments !!! Kill me now ! I'm waiting for the end of august then I'm buying whatever I can get.You might want to make that til Tuesday September 12 when the Paris Apple Expo opens with an Apple keynote.
"Otellini said the company expects to ship its 1 millionth Core 2 Duo processor in a little less than seven weeks after launch. In comparison, it took Pentium a year to reach that level of ramp-up. (http://www.macworld.com/news/2006/07/27/core2duo/index.php)"
Two million cores shipped by this time in September. Wow!
"Otellini said the company expects to ship its 1 millionth Core 2 Duo processor in a little less than seven weeks after launch. In comparison, it took Pentium a year to reach that level of ramp-up. (http://www.macworld.com/news/2006/07/27/core2duo/index.php)"
Two million cores shipped by this time in September. Wow!
FearlessFreep
Apr 11, 01:31 PM
I still don't get why people just don't follow the components to figure out the timing. If the touchscreen size is indeed going to change, then it has to be manufactured in sufficient quantity before launch. Otherwise you end up with not enough product in the pipeline to meet demand (see Ipad, 2).
There's outside factors at work here as well - namely the Japan disaster which has constrained supplies.
Apple may not have any choice but to wait until Fall.
There's outside factors at work here as well - namely the Japan disaster which has constrained supplies.
Apple may not have any choice but to wait until Fall.
CaoCao
Feb 28, 07:35 PM
No, you are absolutely wrong., They can get married like any other couple where the laws allow. Marriage is not a special preserve of any religion. You cannot just commandeer it.
Who cares what Catholic dogma claims? It's an irrelevance.
So what is the problem? Are you against variation?
No, not proof
http://www.fordham.edu/halsall/pwh/1979boswell.html
http://www.newstatesman.com/199908230009
They still can not have valid sacramental marriage
Fornication doesn't matter if the person doesn't care about the religious connotations of marriage
Greek culture also endorsed pederasty!
I wasn't around in the 1970's, but I'm pretty sure that pedophilia wasn't normal then.
Some of this may be media frenzy, but if even one child rapist is hidden by the Catholic Church, it doesn't reflect well on them.
You do realize that the Church did what doctors said to right? They followed the medical professional advice that after treatment they were harmless and could go back.
Now priests can't even defend themselves, now it is guilty until proven innocent, also you don't get a trial to prove the innocence!
Who cares what Catholic dogma claims? It's an irrelevance.
So what is the problem? Are you against variation?
No, not proof
http://www.fordham.edu/halsall/pwh/1979boswell.html
http://www.newstatesman.com/199908230009
They still can not have valid sacramental marriage
Fornication doesn't matter if the person doesn't care about the religious connotations of marriage
Greek culture also endorsed pederasty!
I wasn't around in the 1970's, but I'm pretty sure that pedophilia wasn't normal then.
Some of this may be media frenzy, but if even one child rapist is hidden by the Catholic Church, it doesn't reflect well on them.
You do realize that the Church did what doctors said to right? They followed the medical professional advice that after treatment they were harmless and could go back.
Now priests can't even defend themselves, now it is guilty until proven innocent, also you don't get a trial to prove the innocence!
ChickenSwartz
Aug 5, 10:11 PM
Does anyone think the recent "problems" at Apple are going to have any effect on what happens Monday.
Story: http://www.sfgate.com/cgi-bin/article.cgi?f=/c/a/2006/08/05/BUGAHKBK3H1.DTL
If there are products that are they "maybe" list, this might put them on the "go" list. Big news pushes stock prices up and pushes the "problem" stories on page 2.
Story: http://www.sfgate.com/cgi-bin/article.cgi?f=/c/a/2006/08/05/BUGAHKBK3H1.DTL
If there are products that are they "maybe" list, this might put them on the "go" list. Big news pushes stock prices up and pushes the "problem" stories on page 2.
iJohnHenry
Mar 20, 10:21 AM
This should all go public.
But they have worked so hard, all these decades, to diminish the "one man, one vote" to something much less than that.
Won't you think of their children, and reconsider?
:rolleyes:
But they have worked so hard, all these decades, to diminish the "one man, one vote" to something much less than that.
Won't you think of their children, and reconsider?
:rolleyes:
guzhogi
Jul 20, 10:07 AM
First of all, you assume that it is possible to make "one big core equal in processing power to the 8 cores". I don't think it is possible to do this (at least not with the x86 architecture using today's technology.)
But assuming such a chip exists, the answer depends on what kind of efficiency you're thinking of.
If you mean computational efficiency (meaning the most useful processing per clock-tick), then a single big core will do better. This is because single-threaded apps will be able to use the full power (whereas multiple threads are needed to take advantagte of multiple cores.) Also, the operating system can get rid of the overhead that is needed to keep software running on the multiple cores from stepping on each other.
If you mean energy efficiency (amount of processing per watt of electricity consumed), then it could go either way, depending on how the chips are made. But given today's manufacturing processes and the non-linear power curve that we see as clock speeds are increased, the multiple-core solution will almost definitely use less power.
I remember hearing about how it is possible to make multiple cores act like one (Idon't remember where I heard this). Anyways, whether 8 cores acting separately or together like 1 big processor has an advantage depends on the program you use. If the program is multi-threaded, then the cores acting separately might have the advantage while single threaded apps will have an advantage if the cores are acting like one. However, many apps today won't see that much improvement either way (like a simple calculator, or solitare and word processing).
But assuming such a chip exists, the answer depends on what kind of efficiency you're thinking of.
If you mean computational efficiency (meaning the most useful processing per clock-tick), then a single big core will do better. This is because single-threaded apps will be able to use the full power (whereas multiple threads are needed to take advantagte of multiple cores.) Also, the operating system can get rid of the overhead that is needed to keep software running on the multiple cores from stepping on each other.
If you mean energy efficiency (amount of processing per watt of electricity consumed), then it could go either way, depending on how the chips are made. But given today's manufacturing processes and the non-linear power curve that we see as clock speeds are increased, the multiple-core solution will almost definitely use less power.
I remember hearing about how it is possible to make multiple cores act like one (Idon't remember where I heard this). Anyways, whether 8 cores acting separately or together like 1 big processor has an advantage depends on the program you use. If the program is multi-threaded, then the cores acting separately might have the advantage while single threaded apps will have an advantage if the cores are acting like one. However, many apps today won't see that much improvement either way (like a simple calculator, or solitare and word processing).
mkruck
Apr 6, 04:43 PM
You both ignored HOT DOGS! Sheesh, hot dogs rule. The only problem is kids under 6 choking on them unless you cut them right. But that will be fixed in the v3.0 hot dog, they will come pre-sliced.
Hot dogs?
Hot dogs?
Those are the Hyundai of...oh, wait, that analogy has been used already. Uhhhmmmm, Hot Dogs are the Yugo...oh, man, did it again.
OK, I've got it: Hot Dogs are the Hot Dogs of food.
Hot dogs?
Hot dogs?
Those are the Hyundai of...oh, wait, that analogy has been used already. Uhhhmmmm, Hot Dogs are the Yugo...oh, man, did it again.
OK, I've got it: Hot Dogs are the Hot Dogs of food.
whatever
Jul 20, 03:27 PM
Apple doesn't get to operate in a bubble anymore - in Intel Land you have to compete. If they executed your plan, Dell would just go and make a "workstation" with dual Kentsfield and Apple will get crushed in Photoshop benchmarks. No way they're going to allow that.
Neither Apple or Dell operate in bubbles. They both realize that these chips belong in real servers and also requires an OS that can support such chips.
In 2007 we will finally get to see Apple force their way more into Sun's market.
Neither Apple or Dell operate in bubbles. They both realize that these chips belong in real servers and also requires an OS that can support such chips.
In 2007 we will finally get to see Apple force their way more into Sun's market.
lazyrighteye
Aug 11, 10:50 AM
Using TimeMachine, Steve is going to release it two years ago.
That made my Friday... which may actually be a sad comment on things in my world. :D
That made my Friday... which may actually be a sad comment on things in my world. :D
Mattie Num Nums
Apr 20, 09:57 AM
http://cultofmac.cultofmaccom.netdna-cdn.com/wordpress/wp-content/uploads/2011/04/Screen-shot-2011-04-19-at-8.37.05-PM.png
feel free to point out how difficult it is to see any similarities...
I think 3M should sue apple because the Notes icon looks like a 3M notepad.
feel free to point out how difficult it is to see any similarities...
I think 3M should sue apple because the Notes icon looks like a 3M notepad.
marksman
Mar 22, 03:14 PM
The screen is not 50% smaller. Nice way of making yourself look stupid.
LOL
http://i54.tinypic.com/dma9nn.png
LOL
http://i54.tinypic.com/dma9nn.png
blahblah100
Mar 31, 05:27 PM
The amount of people who never bought an Apple product, but will still log-on to blast away at anything Apple is really quite amusing. ;)
The amount of people who never bought a Microsoft product, but will still log-on to blast away at anything Microsoft is really quite amusing. ;)
Unfortunately, it goes both ways.
The amount of people who never bought a Microsoft product, but will still log-on to blast away at anything Microsoft is really quite amusing. ;)
Unfortunately, it goes both ways.
winterspan
Apr 11, 05:27 PM
80%* of potential purchasers won't have access to LTE for at least another year from then. ...
[*made up statistic, but I bet it's not far wrong! :D ]
Verizon's 4G/LTE service will be live in ~140 metro areas by Q42011
[*made up statistic, but I bet it's not far wrong! :D ]
Verizon's 4G/LTE service will be live in ~140 metro areas by Q42011
aly
Aug 27, 03:58 PM
I'm a bit fed up of all these estimates and random guesses at when apple will release merom updated MBPs. Although I am waiting for the revision before I buy one, I've decided to just wait till I see one instead of getting my hopes up each tuessday :) hehe. Ah well maybe next tuesday.....
boncellis
Jul 20, 12:05 PM
double post, my apologies.
twoodcc
Aug 13, 10:36 PM
You originally said...
...which as I've said, a few times now, is incorrect. If you only count 4 games, as you originally said in that quote, that only totals 46M. And besides, if you check that link I originally provided, which is FROM POLYPHONY THEMSELVES, you would see that the total worldwide is only 56M. At least if you're going to quote the number in the wrong context, use the right number.
well i did have demos just before that quote. but yes, what i meant to say, and what i actually typed, was two different things. but i think you can see what i was trying to say, and is clear: NFS has sold more, but has more games out there.
Yay, let's play the "Twist The Numbers To Fit Our Needs" game....
100M / 15 years = 6.66M per year (Need for Speed)
56M / 13 years = 4.31M per year (Gran Turismo)
So yea, Need for Speed sells 50% more per year. YAY, math is fun!!!
math is fun, and yes, per year, NFS has sold more. but they released more games to do it. and on more consoles, to break it down even further
It is. I don't like the direction Polyphony has taken the game. A game that used to be my favorite game. It disappoints me. Thus my frustration. I bought my PS3 with the hopes
did you buy GT5: prologue?
for someone who at one point really liked the games/series, i don't see why you wouldn't give this game a shot, at least after reading reviews first. i find it very frustrating to see people make judgements before the game is even out
Can I buy one and drive it to work? No? Then it isn't real, intentions or not.
no matter what anyone says, it is a fact that they intended to build a car specifically for Gran Turismo. that in itself is very impressive to me.
Then according to you, Wii Play is the "greatest" video game of all time. Just reading that sentence should show you exactly why sales have very little to do with the quality of a game.
i never said that the most sold game is the greatest. but i do think sales is one aspect to consider.
I keep saying this, but it seems people don't really understand it. Sales, which deal with numbers, are an objective measure of something. Greatness, which deals with personal preferences, is a subjective measure. You can think GT, or the iPhone, or Star Wars, or whatever, is great. That is fine, and a personal opinion. But, the sales of those things can't be "great". They can be large, and they are, but they can't be great.
i understand this, but i don't think you understand what i am trying to point out. you are correct in that what makes something great is personal preference, and when many people have that same personal preference, it turns into many sales.
let's try this analogy. take tennis for example. who is a "greater" player - Roger Federer or Andy Roddick? Roddick has played some great games, and has a great serve, but Federer has many more championships. almost everyone would say Federer is greater (if there is such a term).
think of championships as sales in this case. i know it's not the best analogy, but the first thing i could think of.
And really, if someone uses the sales of something to qualify the greatness of it to themselves, that is kind of sad. Quite a few of my favorite things, which I consider great, didn't sell very well. That doesn't make them any less great to me.
you do make a very good point here, but again, i think you are taking my point out of context.
my whole point in all of this, is not that i think GT is the greatest series of all time or whatever, or that they have sold the most, blah blah. all i'm trying to say is that they have a very respectable racing series, one of the best, and i think this next game will continue their success. and i pointed out that past sales show that they have had much success. that's all i'm really saying here
But anyway, I don't even know why I'm arguing about this. I'm not even going to be buying this game. I'm done with this thread now. I'll just tip my hat, and bid you adieu.
it seems we are arguing just to argue. i hope you at least read some reviews once the game is out before making a choice like that. and i also hope that this thread hasn't led you to that decision
It's refreshing that I don't have to go to gamespot forums to see a pointless immature fanboy pissing match :rolleyes:
sorry for the inconvenience. you don't have to read this you know.
...which as I've said, a few times now, is incorrect. If you only count 4 games, as you originally said in that quote, that only totals 46M. And besides, if you check that link I originally provided, which is FROM POLYPHONY THEMSELVES, you would see that the total worldwide is only 56M. At least if you're going to quote the number in the wrong context, use the right number.
well i did have demos just before that quote. but yes, what i meant to say, and what i actually typed, was two different things. but i think you can see what i was trying to say, and is clear: NFS has sold more, but has more games out there.
Yay, let's play the "Twist The Numbers To Fit Our Needs" game....
100M / 15 years = 6.66M per year (Need for Speed)
56M / 13 years = 4.31M per year (Gran Turismo)
So yea, Need for Speed sells 50% more per year. YAY, math is fun!!!
math is fun, and yes, per year, NFS has sold more. but they released more games to do it. and on more consoles, to break it down even further
It is. I don't like the direction Polyphony has taken the game. A game that used to be my favorite game. It disappoints me. Thus my frustration. I bought my PS3 with the hopes
did you buy GT5: prologue?
for someone who at one point really liked the games/series, i don't see why you wouldn't give this game a shot, at least after reading reviews first. i find it very frustrating to see people make judgements before the game is even out
Can I buy one and drive it to work? No? Then it isn't real, intentions or not.
no matter what anyone says, it is a fact that they intended to build a car specifically for Gran Turismo. that in itself is very impressive to me.
Then according to you, Wii Play is the "greatest" video game of all time. Just reading that sentence should show you exactly why sales have very little to do with the quality of a game.
i never said that the most sold game is the greatest. but i do think sales is one aspect to consider.
I keep saying this, but it seems people don't really understand it. Sales, which deal with numbers, are an objective measure of something. Greatness, which deals with personal preferences, is a subjective measure. You can think GT, or the iPhone, or Star Wars, or whatever, is great. That is fine, and a personal opinion. But, the sales of those things can't be "great". They can be large, and they are, but they can't be great.
i understand this, but i don't think you understand what i am trying to point out. you are correct in that what makes something great is personal preference, and when many people have that same personal preference, it turns into many sales.
let's try this analogy. take tennis for example. who is a "greater" player - Roger Federer or Andy Roddick? Roddick has played some great games, and has a great serve, but Federer has many more championships. almost everyone would say Federer is greater (if there is such a term).
think of championships as sales in this case. i know it's not the best analogy, but the first thing i could think of.
And really, if someone uses the sales of something to qualify the greatness of it to themselves, that is kind of sad. Quite a few of my favorite things, which I consider great, didn't sell very well. That doesn't make them any less great to me.
you do make a very good point here, but again, i think you are taking my point out of context.
my whole point in all of this, is not that i think GT is the greatest series of all time or whatever, or that they have sold the most, blah blah. all i'm trying to say is that they have a very respectable racing series, one of the best, and i think this next game will continue their success. and i pointed out that past sales show that they have had much success. that's all i'm really saying here
But anyway, I don't even know why I'm arguing about this. I'm not even going to be buying this game. I'm done with this thread now. I'll just tip my hat, and bid you adieu.
it seems we are arguing just to argue. i hope you at least read some reviews once the game is out before making a choice like that. and i also hope that this thread hasn't led you to that decision
It's refreshing that I don't have to go to gamespot forums to see a pointless immature fanboy pissing match :rolleyes:
sorry for the inconvenience. you don't have to read this you know.
Fabio_gsilva
Jul 27, 03:47 PM
How about a new Mac at WWDC?
Lower Model:
CConroe E6300 - 1.86 GHz � FSB1066 � 2 MB cache - ($185)
1GB RAM
160GB Serial ATA hard drive
Double-layer SuperDrive (DVD+R DL/DVD�RW/CD-RW)
One open PCI-Express expansion slot
One open Optical drive slot [maybe] (i.e. for 2nd DVD drive)
Graphics Card with 128MB SDRAM
Built-in AirPort Extreme and Bluetooth 2.0, USB/FW800
Remote [(?] I think this box will still be small enough to fit into home entertainment setups.]
Keyboard, Mighty Mouse...................................................... $999
Some Options:
Conroe E6600 - 2.40 GHz � FSB1066 � 4 MB cache � (+$100)
Wireless Keyboard/Mouse +$60
Add DVD/CD ROM drive (in 2nd slot) + $50
250GB SATA hard drive +$75
+1GB RAM (2GB total) +$100
+3GB RAM (4GB total) +$300
Slightly Better Graphics Card with 256MB SDRAM + $50
Much Better Graphics Card +$200+
http://img92.imageshack.us/img92/9648/macandmacminipx9.jpg
Very nice, indeed... (of course, i expect that Apple release something much more beautiful than waht you did :D :p :D)
Lower Model:
CConroe E6300 - 1.86 GHz � FSB1066 � 2 MB cache - ($185)
1GB RAM
160GB Serial ATA hard drive
Double-layer SuperDrive (DVD+R DL/DVD�RW/CD-RW)
One open PCI-Express expansion slot
One open Optical drive slot [maybe] (i.e. for 2nd DVD drive)
Graphics Card with 128MB SDRAM
Built-in AirPort Extreme and Bluetooth 2.0, USB/FW800
Remote [(?] I think this box will still be small enough to fit into home entertainment setups.]
Keyboard, Mighty Mouse...................................................... $999
Some Options:
Conroe E6600 - 2.40 GHz � FSB1066 � 4 MB cache � (+$100)
Wireless Keyboard/Mouse +$60
Add DVD/CD ROM drive (in 2nd slot) + $50
250GB SATA hard drive +$75
+1GB RAM (2GB total) +$100
+3GB RAM (4GB total) +$300
Slightly Better Graphics Card with 256MB SDRAM + $50
Much Better Graphics Card +$200+
http://img92.imageshack.us/img92/9648/macandmacminipx9.jpg
Very nice, indeed... (of course, i expect that Apple release something much more beautiful than waht you did :D :p :D)
greenstork
Aug 17, 05:14 PM
So you have 4hdds in total,with 2 of each in raid 0 or what?
Do you have the os on one pair and scratch on the other pair?
Just out of curiosity, is it even possible to configure a RAID 10 or 01 on OS X setup without a dedicated controller card? I was planning to configure a RAID 1 (two 500 GB drives) on my Mac Pro for the sake of redundancy, but with 4 drives bays to play with, a RAID 10 or 01 might be a little faster if I understand the technology correctly. Anyone?
Do you have the os on one pair and scratch on the other pair?
Just out of curiosity, is it even possible to configure a RAID 10 or 01 on OS X setup without a dedicated controller card? I was planning to configure a RAID 1 (two 500 GB drives) on my Mac Pro for the sake of redundancy, but with 4 drives bays to play with, a RAID 10 or 01 might be a little faster if I understand the technology correctly. Anyone?
dethmaShine
Apr 20, 07:44 AM
No, it was shown at IFA 2006 for the first time but "officially presented" a few months later.
It wasn't shown; it was announced. Get your facts right.
ANd you design hardware in a couple of weeks?
Doesnt really matter LG prada wich has simular concept and specs was designed AND shown over a month before iphone was ever shown.
LG can see the future?
Apple copies, samsung copies, LG copies,... everybody does it only only always wants to make believe everyone copies from them.
No. I don't think that's possible; but samsung didn't ship it until later. So, that could have been a testing unit meant for a future release. Samsung had touch screen plans but could have easily ripped off the design. No big deal there.
It wasn't shown; it was announced. Get your facts right.
ANd you design hardware in a couple of weeks?
Doesnt really matter LG prada wich has simular concept and specs was designed AND shown over a month before iphone was ever shown.
LG can see the future?
Apple copies, samsung copies, LG copies,... everybody does it only only always wants to make believe everyone copies from them.
No. I don't think that's possible; but samsung didn't ship it until later. So, that could have been a testing unit meant for a future release. Samsung had touch screen plans but could have easily ripped off the design. No big deal there.
richard4339
Sep 18, 11:42 PM
I still think it's funny that everyone thinks these Macbook Pros are "long overdue" - when, exactly, did the FIRST Dell laptop with C2D ship? I thought it was supposed to be around tomorrow...but surely it couldn't have been before last Monday or so at the earliest.
So that's, what? A week behind in the worst case scenario? Oh God...
However - if they waited till November, then yeah, I'd agree that they were overdue...:)
Dell has staggered release dates. Today began shipping for the lower end Merom processors (2.0 and 2.16ghz)... the highest end 2.33ghz one was shipping October 1.
So that's, what? A week behind in the worst case scenario? Oh God...
However - if they waited till November, then yeah, I'd agree that they were overdue...:)
Dell has staggered release dates. Today began shipping for the lower end Merom processors (2.0 and 2.16ghz)... the highest end 2.33ghz one was shipping October 1.
gnasher729
Jul 27, 05:59 PM
but is still more productive because it handles more calculations per clock cycle
I'm no processor geek. I have a basic understanding of the terminology and how things work so correct me if I'm wrong, but wasn't this one of the advantages that the PPC had over Intel chips? Does this mean Intel is moving toward shorter pipes? Are we talking more instructions per clock cycle or what? What does "calculations" mean in this context?
With most processors, especially the Intel/AMD processors, "instructions per cycle" is not a useful number. These processors have both simple instructions (add register number 3 to register number 6) and complex instructions (add register number 3 to the number whose address is in register number 6). A PowerPC has the simple instructions, but not the complex ones. Instead it would need three instructions "load the number whose address is in register number 6, and move it to register 7", "add register 3 to register 7", "store register 7 to the location whose address is in register 6". But the Intel processor doesn't magically do three times as much work. Instead, it splits the complex instruction into three so-called "macro-ops", and does exactly the same work. So in this case, the PowerPC would execute three times as many instructions per cycle (3 instead of 1), but because it doesn't do more actual work, that is pointless. Instead you would count the number of operations, and they are more or less the same.
Intel is indeed moving towards shorter pipelines. They have done that already with the Core Duo chips. Longer pipelines have the advantage that each pipeline step is a bit faster, so you can get higher clockspeed. Shorter pipelines have the advantage that they take much less energy (very important; at some point your chips just melt), they are much faster handling branches, and they are just much much easier to design. Pentium 4 needed absolutely heroic efforts to produce it, and would have needed twice the heroics to improve it. Instead, the Core Duo has a much simpler design, that is just as powerful, and because it was so simple, Core 2 Duo could improve it.
And Core 2 Duo can now execute up to four "micro-ops" per cycle, same as the G5, compared to three for Core Duo, Pentium 4 and G4. It also has some clever features that reduce the number of micro-ops needed up to 10 percent, and some other improvements.
I'm no processor geek. I have a basic understanding of the terminology and how things work so correct me if I'm wrong, but wasn't this one of the advantages that the PPC had over Intel chips? Does this mean Intel is moving toward shorter pipes? Are we talking more instructions per clock cycle or what? What does "calculations" mean in this context?
With most processors, especially the Intel/AMD processors, "instructions per cycle" is not a useful number. These processors have both simple instructions (add register number 3 to register number 6) and complex instructions (add register number 3 to the number whose address is in register number 6). A PowerPC has the simple instructions, but not the complex ones. Instead it would need three instructions "load the number whose address is in register number 6, and move it to register 7", "add register 3 to register 7", "store register 7 to the location whose address is in register 6". But the Intel processor doesn't magically do three times as much work. Instead, it splits the complex instruction into three so-called "macro-ops", and does exactly the same work. So in this case, the PowerPC would execute three times as many instructions per cycle (3 instead of 1), but because it doesn't do more actual work, that is pointless. Instead you would count the number of operations, and they are more or less the same.
Intel is indeed moving towards shorter pipelines. They have done that already with the Core Duo chips. Longer pipelines have the advantage that each pipeline step is a bit faster, so you can get higher clockspeed. Shorter pipelines have the advantage that they take much less energy (very important; at some point your chips just melt), they are much faster handling branches, and they are just much much easier to design. Pentium 4 needed absolutely heroic efforts to produce it, and would have needed twice the heroics to improve it. Instead, the Core Duo has a much simpler design, that is just as powerful, and because it was so simple, Core 2 Duo could improve it.
And Core 2 Duo can now execute up to four "micro-ops" per cycle, same as the G5, compared to three for Core Duo, Pentium 4 and G4. It also has some clever features that reduce the number of micro-ops needed up to 10 percent, and some other improvements.
No comments:
Post a Comment