That's simply not the case. Did you ever notice the little barcode imprinted on those envelopes? The postal service scans the bar code and bills the sender for every piece that gets sent. No scan, no pay.
"nice beige"??
hehe.
Sometimes plain and simple is nice... like plain white cotton panties.:-) But I would rather have an interesting looking computer.
AltiVec happens to be the ONLY way to do SIMD type instructions on the PowerPC architechture. The x86 architechture, OTOH, has several incompatible systems: MMX, SSA, SSA-2, 3DNow!, etc. The PowerPC camp will never fall into this problem, since Motorola has liscenced the technology to IBM. And in case you didn't know, only Moto and IBM actually produce PowerPC chips. So, on the PowerPC, it is AltiVec or nothing. And the great thing is, AltiVec kicks the crap out of any one of those SIMD systems on x86.
If you really want to cut out some bloat, start with all those x86 SIMD systems. Maybe support plain-vanilla MMX only or something. But that would just suck, wouldn't it?
Linux users want the best performance possible. AltiVec gives this to them. A few more kB of source to d/l is a small price to pay.
Good point. That would be more efficient that calculating them, huh?
However, wavelets are far more complex, and contain more than a mere 8 values. And (depending on the particular implementation) there can be hundreds or at least dozens of unique wavelets to describe. That makes the implementation slower nonetheless.
AFAIK, Alpha simply does not do OOO. It relies entirely on the compiler to organise the instructions into the most efficient possible order. (almost all compilers for all CPU's attempt to do this to some degree BTW). Leaving out the OOO capability enables the CPU to be much simpler. That lets the designers concentrate on achieving high clock rates. But these simple, fast CPU's are heavily dependant on the compiler delivering efficient optimised code in order to achieve high performance.
Re: MHz differences will fade soon enough...
on
G4 vs. Athlon Review
·
· Score: 3
...As for speed, RISC versus CISC aside, the Motorola/IBM designs have not shown the ability to drive the high clock speeds that Intel and AMD are playing with. Until about a year ago, the two were neck-and-neck, but the X86 chips are now up around 800 MHz while the G4 is just passing 500 now...
True, but the gap will lessen (or disappear) in the near future. The G4 has been limited (in clock speed) by it's exceptionally short 4-stage pipeline. Motorola has demonstrated a version of the G4 with a longer 7-stage pipeline that hits much higher clock speeds (~700 MHz range at the demo - higher in production). Each stage is simpler and faster, resulting in the higher clock speed. The K7 already has a very deep pipeline, which is a large factor in its high clock speed.
Since the K7's FPU handles vector operations, it's not always totally free to do fp ops like the G4's FPU is. But considering that vector and regular fp calculations aren't normally mixed, the K7's fp performance should exceed that of the 7400 under most circumstances...
The G4's vector unit (AlticVec) is way more complex than the K7's. It can do Floating Point operations - four SP (single precision) or two DP (double prec) in fact. In combination with the FPU of the G4 (which can do one SP or DP FP op), the G4 can do no fewer than five SP FP ops or three DP FP ops per cycle. Any application that does FP ops and is compiled on an AtliVec enabled compiler (such as Codewarrior or Mototrola's) will take advantage of this superior capability. AltiVec's 32, 128 bit-wide vector registers and it's 155 vector instructions make it a formidable number-cruncher.
IMHO, the PowerPC has more "room to grow" than any x86 chip, whether Intel, AMD, or otherwise. The reason is this: Power Consumption.
IBM and Motorola are working on putting multiple PowerPC cores onto a single die. IBM has already done this with it's Power CPU's (a sibling to the PowerPC). This is feasible with the PowerPC, since its power consumption is so very low. A G3 at 400 MHz (on.22 micron process) for example uses 8 W max, 5 W on average. A single PIII or Athlon uses at least 4-5 times that much on average. This is due in large part to the complex instruction set that must be decoded and executed. With IBM's and Moto's superior Copper interconnect and SOI technologies, the power comsumption and core size can be reduced further, allowing even more cores on each die. Modern Multitasking, multiprocessing OS's with well written multithreaded apps will scream on these multiple-core CPU's.
3. How CPU intensive is it to decode these things? Will MJPEG2000 (or whatever) practically require a hardware decoder for DVD-quality playback?
Much more CPU intensive than JPEG or MPEG, which use DCT's (Discrete Cosine Transforms) because CPU's have instructions to generate sine or cosine waves, but none to generate wavelets. Each wavelet in the image must be generated on-the-fly or looked up in a table, both of which would take more of hit on the CPU.
Re:When will age of single monolithic CPU end?
on
News on Pentium IV
·
· Score: 1
What's really needed is well-written, multi-threaded applications. That way, separate threads can be executed by any CPU in the system (provided of course you have a SMP aware kernel).
The Mach microkernel is a great foundation for an efficient SMP O/S. There is a version of Linux that runs on top of Mach already... has been for years now. It's called MkLinux and was created and is funded by Apple (hooray, Apple!). Versions now exist for both PowerPC (Macs) and intel x86 based machines. Apple has taken their knowledge of Mach and is using it in their next-gen O/S, Mac OS X (pronounced "ten"), which will really rock!
Re:When will age of single monolithic CPU end?
on
News on Pentium IV
·
· Score: 1
Hmm. That's an interesting idea. IBM is doing that now, with their Power line of CPU's (not yet with PowerPC, unfortunately)
But then, IBM can do this b/c their CPU's only use a few watts of power each and are only a few mm in size. Putting serveral on a single die is doable. Intel PII/PIII's OTOH are several times as big and several times as hot. Putting multiple PIII cores on a single die (with current manufacturing processes) is laughable.
The problem is not that they are doing legitamate business, but that they are trying to prevent other online companies from doing legitamate business!!
Amazon did not originate the ideas protected by their patent. They merely were the first ones to have the gall to try to patent them. And they got the damned patent!! (WTF?)
Here, Here!! Unfortunately, there is no way of executing such a grand plan. Absolutely anyone can set up a mail server on their own computer, so there really is no way of measuring the quantity of e-mail that a person sends -- appart from simple raw IP traffic measurement, but that could be any type of traffic.
Blank CD-R's (not CD-RW's) were $10 each in 1997 - you don't have to look back as far as 1991. Then you could find them for $5 if you were lucky. Then there was the period of rebates, where you got back most of the purchase price - almost every manufacturer offered them. Now they are just $1-$1.50 a disc. The RIAA is the one that should be scared - 12-15 albums worth of mp3's fit on a $1 disc.
Steve Jobs did save Apple, no doubt about it. And he did it by cancelling projects that didn't make money. Apple had a lot of really really cool technology before Jobs. Cyberdog, Darwin, OpenDoc... all of it got "Steved". SGI needs to do the same thing to save itself. FOCUS. Now Apple is focused on producing great professional computers (G4's, PowerBooks), great consumer computers (iMac's, iBooks) and a great OS (Mac OS X a.k.a. Rhapsody). They even open-sourced the core technology of Mac OS X. Sweet. SGI needs to focus on high-end workstations, Irix, and MIPS. And they could join the party and open-source openGL, which would be cool of them.
You're confusing a power plug with a power jack. The wall "brick" that you buy from Radio Shack has a plug on it that plugs into the jack on the lego rx. Unfortunately, the new 1.5 version has no jack.
The original poster mentioned considering a RAID level 5 array to try to speed up access. However, Level 5 can actually slow down access times. It increases throughput, but throughput is usually not the culprit. Access time is. To get faster access times, use a mirrored array (level 1), where multiple disks all carry identical information. Read access times are dramatically improved, because each disk can service just a fraction of the overall read requests!! In such an array, reads don't involve all disks, only writes do. Therefore, doubling the number of disks in a mirrored arrar theoretically doubles the number of read transactions that can by done per second. Real-world results vary, but are dramatically better than with a single disk. If the disk is getting a lot of small read transactions per second, rather than a few very large ones, then a mirrored array is the way to go, not striped!
Where MD really shines is in the portables. You can get portable recorders that are barely bigger than the discs (2.5 inch square) themselves and high bitrate) MP3's.
CD-R is great. The media is certainly cheaper than MD. CD sounds better than MD. But it is record-once, and portable playback units, by necessity of the media size, are about 4 times as big as MD units. My MD player runs 12 hrs. on a single AA, it displays album and song titles, and it fits in my shirt pocket. I like it.
Betamax was DOA. MD, however, is extremely poplular in Japan, with sales of pre-recorded MD's outpacing even CD's. That ensures a long life even here, where it is less popular. Just my.02
BTW: DVD isn't (technically) going to look a whole lot better on HDTV. It doesn't have enough resolution lines to make use of it... Buuut, the increased width of the display will allow for a few more scanlines. Not enough to make a big deal over though. I think it will look better though, due to tube/flat screen/lcd/etc.. technology advances giving a sharper image. Not due to the HDTV standard, though.
Yes, it will look better in a HDTV set. Most HDTV sets will double the number of scanlines of the NTSC signal, resulting in a less flickery picture without noticable scanlines. And if you buy a DVD player with progressive component video outputs, then it will look superb, without the interlacing effects ("combing") and "jutter" caused by converting a 24 frame/sec signal to 60 field/sec.
Memory Protection has nothing to do with the user level of a given process (program as you called it). The kernel determines whether the code being executed currently is running in user mode or root user mode or supervisor (kernel) mode. It has nothing to do with the memory mapping. Read a book or two.
Slashdot Mirror
That's simply not the case. Did you ever notice the little barcode imprinted on those envelopes? The postal service scans the bar code and bills the sender for every piece that gets sent. No scan, no pay.
"nice beige"?? hehe. Sometimes plain and simple is nice... like plain white cotton panties. :-) But I would rather have an interesting looking computer.
You are soooo wrong.
AltiVec happens to be the ONLY way to do SIMD type instructions on the PowerPC architechture. The x86 architechture, OTOH, has several incompatible systems: MMX, SSA, SSA-2, 3DNow!, etc. The PowerPC camp will never fall into this problem, since Motorola has liscenced the technology to IBM. And in case you didn't know, only Moto and IBM actually produce PowerPC chips. So, on the PowerPC, it is AltiVec or nothing. And the great thing is, AltiVec kicks the crap out of any one of those SIMD systems on x86.
If you really want to cut out some bloat, start with all those x86 SIMD systems. Maybe support plain-vanilla MMX only or something. But that would just suck, wouldn't it?
Linux users want the best performance possible. AltiVec gives this to them. A few more kB of source to d/l is a small price to pay.
> *drool*
I agree! That's some nice eye-candy. Or should I say iCandy?
However, wavelets are far more complex, and contain more than a mere 8 values. And (depending on the particular implementation) there can be hundreds or at least dozens of unique wavelets to describe. That makes the implementation slower nonetheless.
AFAIK, Alpha simply does not do OOO. It relies entirely on the compiler to organise the instructions into the most efficient possible order. (almost all compilers for all CPU's attempt to do this to some degree BTW). Leaving out the OOO capability enables the CPU to be much simpler. That lets the designers concentrate on achieving high clock rates. But these simple, fast CPU's are heavily dependant on the compiler delivering efficient optimised code in order to achieve high performance.
True, but the gap will lessen (or disappear) in the near future. The G4 has been limited (in clock speed) by it's exceptionally short 4-stage pipeline. Motorola has demonstrated a version of the G4 with a longer 7-stage pipeline that hits much higher clock speeds (~700 MHz range at the demo - higher in production). Each stage is simpler and faster, resulting in the higher clock speed. The K7 already has a very deep pipeline, which is a large factor in its high clock speed.
IBM and Motorola are working on putting multiple PowerPC cores onto a single die. IBM has already done this with it's Power CPU's (a sibling to the PowerPC). This is feasible with the PowerPC, since its power consumption is so very low. A G3 at 400 MHz (on .22 micron process) for example uses 8 W max, 5 W on average. A single PIII or Athlon uses at least 4-5 times that much on average. This is due in large part to the complex instruction set that must be decoded and executed. With IBM's and Moto's superior Copper interconnect and SOI technologies, the power comsumption and core size can be reduced further, allowing even more cores on each die. Modern Multitasking, multiprocessing OS's with well written multithreaded apps will scream on these multiple-core CPU's.
Much more CPU intensive than JPEG or MPEG, which use DCT's (Discrete Cosine Transforms) because CPU's have instructions to generate sine or cosine waves, but none to generate wavelets. Each wavelet in the image must be generated on-the-fly or looked up in a table, both of which would take more of hit on the CPU.
The Mach microkernel is a great foundation for an efficient SMP O/S. There is a version of Linux that runs on top of Mach already... has been for years now. It's called MkLinux and was created and is funded by Apple (hooray, Apple!). Versions now exist for both PowerPC (Macs) and intel x86 based machines. Apple has taken their knowledge of Mach and is using it in their next-gen O/S, Mac OS X (pronounced "ten"), which will really rock!
But then, IBM can do this b/c their CPU's only use a few watts of power each and are only a few mm in size. Putting serveral on a single die is doable. Intel PII/PIII's OTOH are several times as big and several times as hot. Putting multiple PIII cores on a single die (with current manufacturing processes) is laughable.
The problem is not that they are doing legitamate business, but that they are trying to prevent other online companies from doing legitamate business!!
Amazon did not originate the ideas protected by their patent. They merely were the first ones to have the gall to try to patent them. And they got the damned patent!! (WTF?)
For a list of accredited alternatives to the NSI , check this out
Here, Here!
For a list of alternative domain registars other the NSI, check this out.
Here, Here!! Unfortunately, there is no way of executing such a grand plan. Absolutely anyone can set up a mail server on their own computer, so there really is no way of measuring the quantity of e-mail that a person sends -- appart from simple raw IP traffic measurement, but that could be any type of traffic.
Blank CD-R's (not CD-RW's) were $10 each in 1997 - you don't have to look back as far as 1991. Then you could find them for $5 if you were lucky. Then there was the period of rebates, where you got back most of the purchase price - almost every manufacturer offered them. Now they are just $1-$1.50 a disc. The RIAA is the one that should be scared - 12-15 albums worth of mp3's fit on a $1 disc.
Mmmm.... candy colored computers.
Steve Jobs did save Apple, no doubt about it. And he did it by cancelling projects that didn't make money. Apple had a lot of really really cool technology before Jobs. Cyberdog, Darwin, OpenDoc... all of it got "Steved". SGI needs to do the same thing to save itself. FOCUS. Now Apple is focused on producing great professional computers (G4's, PowerBooks), great consumer computers (iMac's, iBooks) and a great OS (Mac OS X a.k.a. Rhapsody). They even open-sourced the core technology of Mac OS X. Sweet. SGI needs to focus on high-end workstations, Irix, and MIPS. And they could join the party and open-source openGL, which would be cool of them.
--
MacBoy
that's my story and I'm stickin' to it
You're confusing a power plug with a power jack. The wall "brick" that you buy from Radio Shack has a plug on it that plugs into the jack on the lego rx. Unfortunately, the new 1.5 version has no jack.
The original poster mentioned considering a RAID level 5 array to try to speed up access. However, Level 5 can actually slow down access times . It increases throughput, but throughput is usually not the culprit. Access time is. To get faster access times, use a mirrored array (level 1), where multiple disks all carry identical information. Read access times are dramatically improved, because each disk can service just a fraction of the overall read requests!! In such an array, reads don't involve all disks, only writes do. Therefore, doubling the number of disks in a mirrored arrar theoretically doubles the number of read transactions that can by done per second. Real-world results vary, but are dramatically better than with a single disk. If the disk is getting a lot of small read transactions per second, rather than a few very large ones, then a mirrored array is the way to go, not striped!
Where MD really shines is in the portables. You can get portable recorders that are barely bigger than the discs (2.5 inch square) themselves and high bitrate) MP3's.
CD-R is great. The media is certainly cheaper than MD. CD sounds better than MD. But it is record-once, and portable playback units, by necessity of the media size, are about 4 times as big as MD units. My MD player runs 12 hrs. on a single AA, it displays album and song titles, and it fits in my shirt pocket. I like it.
Betamax was DOA. MD, however, is extremely poplular in Japan, with sales of pre-recorded MD's outpacing even CD's. That ensures a long life even here, where it is less popular. Just my .02
10 or 20 points? That's laughable. It will drop a lot more than that.
Memory Protection has nothing to do with the user level of a given process (program as you called it). The kernel determines whether the code being executed currently is running in user mode or root user mode or supervisor (kernel) mode. It has nothing to do with the memory mapping. Read a book or two.
Their page loaded very quickly for me.
Maybe the Mac webserver is doing a good job :-)