SCO reported declines in product and services revenue in the six months ending April 2003 compared to the same period last year. However, those declines were offset by $8 in new licensing revenues.
Whoa! No wonder they're so desperate for new licensing revenue.:)
Err, maybe you should have looked up democracy too:
1 a : government by the people; especially : rule of the majority b: a government in which the supreme power is vested in the people and exercised by them directly or indirectly through a system of representation usually involving periodically held free elections 2: a political unit that has a democratic government 3capitalized: the principles and policies of the Democratic party in the U.S. 4: the common people especially when constituting the source of political authority 5: the absence of hereditary or arbitrary class distinctions or privileges
You see, in our democracy, while we have representatives to take care of the bulk of issues, we still have free elections where we all vote on propositions, measures and those representatives.
A 15k rpm drive can read 500KB in the time it takes to seek to a new piece of data. Now what do you think happens more frequently while you use a computer, seeking to new locations on the disc, or reading contiguous blocks of 500KB or more? Now, unless you are streaming massive amounts of sequential data (eg. HD Video), your bottleneck will be access time, not throughput.
The rotational speed of the drive is directly related to the access time. If the data you want is on the other side of the platter, you must wait for it to rotate 180 degrees before you can start reading, regardless of whether the disc is 1/2" in diameter or 2" diameter, whether there are 1GB per square inch or 10GB per square inch.
When the head gets lined up with the track and ready to read, the data it's waiting for can be anywhere between 0 degrees and 360 degrees away. If you average out all those possibilities, you can expect the data to be about 180 degrees away.
Now, a 15000 rpm drive rotates 180 degrees 30000 times per minute. Conversely, it takes 2ms to rotate 180 degrees. If you consider that a typical 15k rpm drive has an average seek time of 3.3ms and we know that 2ms are spent waiting for the disk to spin, than 1.3ms must be spent moving the head. This proves to me that rotational speed is more important to access time than data density.
I'm no hard drive engineer, but I would bet that an increase in density would mean a decrease for rotational speed since a read head probably has a limited bandwidth. (This is probably why the faster-spinning drives typically hold less data.) If you halve the time moving the head while doubling the time waiting for spinning data, you will see an overall increase in seek time.
My conclusion is that greater density and less rpms would hurt access time which is the most important performance factor. However, like the "MHz myth", I'm sure marketing will focus on bandwidth benchmarks for performance instead of real-life application performance.
I'm a little confused as to why the article just references 'QuickTime' when really we're talking about only one codex out of the (what, 200?) media types QT understands.
They call it the "QuickTime" AAC codec because Apple distributes it within QuickTime. It would, however, be less confusing if they referred to it as the "Apple" AAC codec, since you could theoretically add any number of 3rd party AAC codecs to QuickTime. Inside QuickTime, this codec is listed as "MPEG-4 Audio."
What I'm curious about is, there was some discussion before about the differences between the original AAC encoder that came with QT, and the newer one that now ships with QT 6.3 (and ties to iTunes). The original encoder was said to have sucked. This one, if I'm reading this right, is now very good...?
Apple's first release of their AAC codec was not optimal and suffered from the much-discussed quality problems. They put a lot of work into improving it for QuickTime 6.3 which was directly tied to their release of iTunes 4 and the Apple Music Store. Obviously, they needed a good, high-quality codec if people were to take the Music Store seriously. We are simply dealing with an upgraded version of that same AAC codec.
Troll1: First Blot!
Troll2: It's that dude in the goatse picture!
Troll3: A slashdot poster who can't spell.
Troll4: It's a duplicate of the second blot!
That's true, but only if you don't recompile your applications specifically for the G5. If you rebuild (or do a separate build) with the compiler option -mpowerpc64, you get LP64 (longs and pointers are 64-bit, in other words). When you do that, you can malloc to your heart's content, up to 18 billion gigs of virtual memory. (You will run out of swap long before that, of course.)
That's what I thought too when he told me about the limitation. I asked him if recompiling for G5 would allow my program to access past the 32-bit address barrier and he said "no, you are still limited to 4GB."
The Apple engineer I worked with in the G5 performance lab at WWDC told me that while the OS can address over 4 GB of memory using 64-bit addresses, each running process was limited to a 4 GB virtual address space.
I think if you ask most customers, they'ld prefer software that works and solves their problem.
I think you are confusing customers with users. I'm sure the end-users would much rather have the reliable software. But all too often, those users are not the customers who make the purchasing decisions. Those with the decision-making power are reading marketing data-sheets, listening to sales-pitches and browsing product web-sites, forming their opinions based on supposed features, not from experience with the product.
You say it's the "fast" USB, but I'm still confused--maybe I'm just an idiot. Is it the "high-speed" or the "full-speed" version? I mean... they both sound fast.
Re:This will be another solid update
on
Jaguar is Over
·
· Score: 2, Interesting
It's a beautiful thing, because you can still use OS 10.0 if you want to, but they add so many features, bells, whistles and in general cool stuff - people really want to get the newest version of their software.
I wish it were that simple. If you want to run the latest applications, then no, you can't use OS 10.0. Most apps require 10.2, and almost all others require 10.1.
Everyone is not able to buy their way into the carpool lane. They propose to auction off a limited number of passes on eBay. Through the auctioning process, they can see how high the stickers are bid up - and therefore, what people are willing to pay for access to the fast lanes.
Once they understand what this is worth to commuters, they can regulate the toll price to keep the lanes running smoothly.
Are they actually saying that someone inducing thought into their culture from the west might cause an uproar?
*Gasp*
That questioning the truth is a bad thing?
Please, don't call "The Truth" whatever is discussed in the Matrix : you've got your perception of the reality whereas Egyptians may have theirs.
Re-read that comment. Where did the above poster call "The Trush" whatever is discussed in the Matrix? They are not saying that the Matrix suggests a "truth", but that it merely calls "truth" into question--something which may inspire thinking for one's self.
It's sad that governments do this, but not surprising. I imagine it's much easier to maintain control over people who don't try to think for themselves. Whether you like it or not, it's the government's path of least resistance.
At $5/MB after I use up the first 5MB, that spam would cost ME $0.30! It is for this reason that I am NOT going to subscribe to a wireless internet service yet. I was seriously considering it the other day, but then realized I really do not have 100% control over what comes down that pipe at my expense.
Now take it a step further... what if I flushed my IMAP cache for some reason? I could easily pay another $0.30 to download that same spam!
There's no satisfaction in multi-player gaming like turning the tables on the cheaters...
Years ago, when QuakeWorld was popular, some friends and I were running a QW server. Skins were a new feature, so we installed the latest skin packs on our server and made use of them. At one point, we noticed that some players were using skins that were totally black or near-black camoflage so that they were virtually invisible in most areas of the maps. So I had a little idea...
I went through the skin packs and found all of these 'cheater' skins. Then I replaced each one with a custom skin I had made. It was 100% glowing white with a big glowing red target on the front and read "Kick Me" on the back. I passed the updated cheater skins to all of my friends and installed them on the server. Now, all new players, my friends, and I would see the glowing skins while the cheaters themselves thought they were using the black skins.
It was beautiful! They were so easy to spot and aim at.:) 90% of the time you'd see these glowing-white shapes hiding out in the deep dark shadows of the map camping out for us. Hiding in the corners, glowing white, they were just that much easier to blow away with rockets. They would say things like "cheaterbob> How did you see me?!", and "cheaterjoe> You guys must be cheating!".
Some dude was fired from our company a while back.
When we sorted through his equipment, not only did he have volumes of she-male pr0n, but he had been subscribed to she-male pr0n emailing lists using his company email account.
It certainly explained his freaky looking "girlfriend". :)
I don't care how fat my HD pipe is. I just want it to be responsive!
How many of us are pushing huge streams of data in and out of our drives? Unless you're working with uncompressed video, I doubt many.
How much longer must I wait for 10k or 15k IDE drives?
Does anyone know how to measure whether your system is spending more time finding data or pushing data?
If software initiates a read request at time t0, the drive begins reading at t1, and the request is completed at t2, what is larger on average? (t1-t0) or (t2-t1)?
How many developers do you think waste their time programming AltiVec in assembly?
You can accomplish 95% of your optimization by using compiler extensions which expose AltiVec & MMX/SSE/SSE2 as intrinsic C functions. I code with variables, not registers and I let the compiler worry about instruction scheduling and register contention. This is incredibly easy to implement and maintain, which is not something that could be said of assembly. If you want to squeeze out that last 5%, then go ahead and do it in assembly. Meanwhile, I'll be busy optimizing the next algorithm.
As for that fully tuned AltiVec iDCT routine, I'd point you towards Intel's website, where you can download and use such common routines as DCT and iDCT. Intel has a vested interest in having your apps run fast on their processors.
As for vec_permute(), I agree that it's a cool instruction, but one of its most common uses is to work around the AltiVec limitation that prevents you from loading or storing data to unaligned memory locations. Not an issue on Pentiums.
I'll believe Apple has an x86 port of OS X. But, the OS is not the main problem. Apple can't switch to the x86 because their app vendors would be in hell trying to port their AltiVec optimizations to MMX/SSE/3DNow!/SSE2. Nevermind that Apple would suddenly be just another PC vendor.
I'm sorry, but have you ever written code and optimized it with both AltiVec and SSE2?
I have and I can tell you that the bulk of porting between the two is "Copy", "Paste", and "Search and Replace".
For example:, you might replace 'vec_add' with '_mm_add_epi16', 'vec_pack' with '_mm_packus_epi16', or 'vector signed short' with '__m128i'.
This goes for about 90% of my SIMD code. The last 10% is where the two instruction sets do not exactly match up: A simple assignment in AltiVec is an _mm_load_si128() in SSE2. The different byte-orders sometimes introduces a few changes. Some instructions behave differently, eg. vec_madd() and _mm_madd_epi16().
In the end, porting an optimization between AltiVec and SSE2 is trivial compared to writing the optimized algorithm in the first place.
SSE: It's no different from SSE2 if you're just doing single-precision floating-point.
MMX: It's the same story, except you also have to cut your 16-byte vectors down to 8-bytes.
3DNow!: Again, same idea, but with a much smaller palette of instructions to choose from.
Slashdot Mirror
Err, maybe you should have looked up democracy too:
You see, in our democracy, while we have representatives to take care of the bulk of issues, we still have free elections where we all vote on propositions, measures and those representatives.Geez
The rotational speed of the drive is directly related to the access time. If the data you want is on the other side of the platter, you must wait for it to rotate 180 degrees before you can start reading, regardless of whether the disc is 1/2" in diameter or 2" diameter, whether there are 1GB per square inch or 10GB per square inch.
When the head gets lined up with the track and ready to read, the data it's waiting for can be anywhere between 0 degrees and 360 degrees away. If you average out all those possibilities, you can expect the data to be about 180 degrees away.
Now, a 15000 rpm drive rotates 180 degrees 30000 times per minute. Conversely, it takes 2ms to rotate 180 degrees. If you consider that a typical 15k rpm drive has an average seek time of 3.3ms and we know that 2ms are spent waiting for the disk to spin, than 1.3ms must be spent moving the head. This proves to me that rotational speed is more important to access time than data density.
I'm no hard drive engineer, but I would bet that an increase in density would mean a decrease for rotational speed since a read head probably has a limited bandwidth. (This is probably why the faster-spinning drives typically hold less data.) If you halve the time moving the head while doubling the time waiting for spinning data, you will see an overall increase in seek time.
My conclusion is that greater density and less rpms would hurt access time which is the most important performance factor. However, like the "MHz myth", I'm sure marketing will focus on bandwidth benchmarks for performance instead of real-life application performance.
They call it the "QuickTime" AAC codec because Apple distributes it within QuickTime. It would, however, be less confusing if they referred to it as the "Apple" AAC codec, since you could theoretically add any number of 3rd party AAC codecs to QuickTime. Inside QuickTime, this codec is listed as "MPEG-4 Audio."
What I'm curious about is, there was some discussion before about the differences between the original AAC encoder that came with QT, and the newer one that now ships with QT 6.3 (and ties to iTunes). The original encoder was said to have sucked. This one, if I'm reading this right, is now very good...?
Apple's first release of their AAC codec was not optimal and suffered from the much-discussed quality problems. They put a lot of work into improving it for QuickTime 6.3 which was directly tied to their release of iTunes 4 and the Apple Music Store. Obviously, they needed a good, high-quality codec if people were to take the Music Store seriously. We are simply dealing with an upgraded version of that same AAC codec.
Troll1: First Blot!
Troll2: It's that dude in the goatse picture!
Troll3: A slashdot poster who can't spell.
Troll4: It's a duplicate of the second blot!
this could go on forever... :)
That's what I thought too when he told me about the limitation. I asked him if recompiling for G5 would allow my program to access past the 32-bit address barrier and he said "no, you are still limited to 4GB."
That sounds rather limited to me.
I think you are confusing customers with users. I'm sure the end-users would much rather have the reliable software. But all too often, those users are not the customers who make the purchasing decisions. Those with the decision-making power are reading marketing data-sheets, listening to sales-pitches and browsing product web-sites, forming their opinions based on supposed features, not from experience with the product.
You say it's the "fast" USB, but I'm still confused--maybe I'm just an idiot. Is it the "high-speed" or the "full-speed" version? I mean... they both sound fast.
I wish it were that simple. If you want to run the latest applications, then no, you can't use OS 10.0. Most apps require 10.2, and almost all others require 10.1.
Once they understand what this is worth to commuters, they can regulate the toll price to keep the lanes running smoothly.
Re-read that comment. Where did the above poster call "The Trush" whatever is discussed in the Matrix? They are not saying that the Matrix suggests a "truth", but that it merely calls "truth" into question--something which may inspire thinking for one's self.
It's sad that governments do this, but not surprising. I imagine it's much easier to maintain control over people who don't try to think for themselves. Whether you like it or not, it's the government's path of least resistance.
I sure hope you couldn't tell them apart---PNG is lossless compression. :)
I just looked--I have some spams in my inbox upwards of 60KB.
Now suppose for a moment that I had signed up for T-Mobile's $20/month Wireless Internet service plan.
At $5/MB after I use up the first 5MB, that spam would cost ME $0.30! It is for this reason that I am NOT going to subscribe to a wireless internet service yet. I was seriously considering it the other day, but then realized I really do not have 100% control over what comes down that pipe at my expense.
Now take it a step further... what if I flushed my IMAP cache for some reason? I could easily pay another $0.30 to download that same spam!
Years ago, when QuakeWorld was popular, some friends and I were running a QW server. Skins were a new feature, so we installed the latest skin packs on our server and made use of them. At one point, we noticed that some players were using skins that were totally black or near-black camoflage so that they were virtually invisible in most areas of the maps. So I had a little idea...
I went through the skin packs and found all of these 'cheater' skins. Then I replaced each one with a custom skin I had made. It was 100% glowing white with a big glowing red target on the front and read "Kick Me" on the back. I passed the updated cheater skins to all of my friends and installed them on the server. Now, all new players, my friends, and I would see the glowing skins while the cheaters themselves thought they were using the black skins.
It was beautiful! They were so easy to spot and aim at. :) 90% of the time you'd see these glowing-white shapes hiding out in the deep dark shadows of the map camping out for us. Hiding in the corners, glowing white, they were just that much easier to blow away with rockets. They would say things like "cheaterbob> How did you see me?!", and "cheaterjoe> You guys must be cheating!".
Sweet, sweet revenge...
When we sorted through his equipment, not only did he have volumes of she-male pr0n, but he had been subscribed to she-male pr0n emailing lists using his company email account.
It certainly explained his freaky looking "girlfriend".
:)
Maybe you're thinking of RAID 5 where you need 4 drives (I think) to get both redundancy and performance.
How many of us are pushing huge streams of data in and out of our drives? Unless you're working with uncompressed video, I doubt many.
How much longer must I wait for 10k or 15k IDE drives?
Does anyone know how to measure whether your system is spending more time finding data or pushing data?
If software initiates a read request at time t0, the drive begins reading at t1, and the request is completed at t2, what is larger on average? (t1-t0) or (t2-t1)?
All these people were so freaked out by what they thought he could do with a payphone, they denied him his every right.
It's because the RIAA got wind of Saddam Hussein's pirated music archive!
Here's a better idea: Why don't these anal SOB's place their ads inline!
Go fix it dear Henry, dear Henry, dear Henry
Go fix it dear Henry, dear Henry, fix it.
Sorry, but I just couldn't get that song out of my head after reading that. :)
You can accomplish 95% of your optimization by using compiler extensions which expose AltiVec & MMX/SSE/SSE2 as intrinsic C functions. I code with variables, not registers and I let the compiler worry about instruction scheduling and register contention. This is incredibly easy to implement and maintain, which is not something that could be said of assembly. If you want to squeeze out that last 5%, then go ahead and do it in assembly. Meanwhile, I'll be busy optimizing the next algorithm.
As for that fully tuned AltiVec iDCT routine, I'd point you towards Intel's website, where you can download and use such common routines as DCT and iDCT. Intel has a vested interest in having your apps run fast on their processors.
As for vec_permute(), I agree that it's a cool instruction, but one of its most common uses is to work around the AltiVec limitation that prevents you from loading or storing data to unaligned memory locations. Not an issue on Pentiums.
I have and I can tell you that the bulk of porting between the two is "Copy", "Paste", and "Search and Replace".
For example:, you might replace 'vec_add' with '_mm_add_epi16', 'vec_pack' with '_mm_packus_epi16', or 'vector signed short' with '__m128i'.
This goes for about 90% of my SIMD code. The last 10% is where the two instruction sets do not exactly match up: A simple assignment in AltiVec is an _mm_load_si128() in SSE2. The different byte-orders sometimes introduces a few changes. Some instructions behave differently, eg. vec_madd() and _mm_madd_epi16().
In the end, porting an optimization between AltiVec and SSE2 is trivial compared to writing the optimized algorithm in the first place.
SSE: It's no different from SSE2 if you're just doing single-precision floating-point.
MMX: It's the same story, except you also have to cut your 16-byte vectors down to 8-bytes.
3DNow!: Again, same idea, but with a much smaller palette of instructions to choose from.