All the systems for the SPEC numbers I showed use the same number of processors.
I think you miss the point. SPEC isn't necessarily telling for real world speeds - especially on dual processor units. I agree SPEC is important. But it certainly isn't the whole story.
Actually according to all the benchmarks of Mathematica 5 I've seen, the top end x86 beats the dual G5 there as well. I'd expect the Athalon-64 to win handedly there as well. The G5's poor Integer performance really cripples it in those sorts of benchmarks. Which isn't to say it isn't extremely fast. Just that for many typical applications it is slower than top end PC's (although somewhat compatible in general speed). Of course then the fair comment is the price difference between the top PCs and the top Mac. And there is a huge price difference.
We'll see what happens. Competition is good for everyone. I prefer, for a variety of reasons, OSX. And by and large I think performance gets too much focus by people who will never use it. But at the same time iMovie and FCE shows digital editing is coming to the masses and something faster than even a G5 or Athalon-64 would be nice there.
But I agree that Mathematica would be a better choice. The problem with the test was that it was confused about whether it was testing platforms to do common tasks or simply comparing the speed of the platforms. For the former then it really should have used better applications on the Mac, such as FCP. For the latter it also chose a rather questionable suite.
Those numbers do speak for the chip. And I don't think most Mac users would debate it. However this is a dual processor versus a single processor. So this isn't entirely a fair comparison for the systems.
Further I think the criticisms of the applications tested is entirely appropriate. It really was a very questionable test. I think the Athalon-64 would win a fair test as well. Just that this wasn't that test.
Anyone who thinks they'll get better Officer performance on either XP or OSX with the new chips are right, but wrong at the same time. A 500 MHz PIII runs OfficeXP handily with no speed perception problems. Anything else, unless you are really hitting complex Excel worksheets (doubtful), is overkill. Same on the Mac, albeit slightly less so due to Microsoft's code. But lets say a 800 MHz G4 to be fully satisfied.
For the vast majority of users faster chips are overkill. They are of benefit to people doing complex graphics, scientific programming, databases, and games. And yes, that latter is what counts to many people. I personally consider it idiocy what people spend to play video games, but that's me. But for probably 80% of the people using computers what came out two years ago was more than powerful enough.
To do benchmarks on such programs seems silly to me.
Anyone buying an Athalon-64 primarily to run the latest incarnation of Doom or Office is simply an idiot. If they happen to run those as well, that's fine.
And, as an other poster pointed out, overall the suite of products used to test the computers makes little sense. I mean I expect the Athalon to be faster. Always did. But if you want to make a meaningful comparison use applications that make some sense.
Panther isn't a full 64-bit OS. However it does have math and other libraries optimized for 64-bit instructions and has a library for addressing more than 4 GB of RAM. It's process manager can address more than 4 GB total, with 4 GB max going to each process. Although more can be addressed indirectly via the use of the afore mentioned libraries.
Having said that few people need a full 64-bit OS and Panther is aimed squarely at consumers. I expect 64 bit pointers to come eventually, but it is probably better to stick with 32 bit pointers and keep better compatibility.
And, let's be honest. Most people running the Athalon-64 will be using it just as a fast Athalon running WindowsXP.
MacOS offers no significant value over and above Linux.
I'd reject that assumption for many applications.
The desktop of OSX is vastly superior to Linux on the Desktop of even products like Ximian. Further there is far more productivity software for OSX than Linux.
Realistically though one can ask if a x86 OSX would have all its software ported. So you do have a point. But I think that in terms of user friendliness OSX beats any Linux distro and saying that Linux will get there "real soon now" is still wishful thinking. (IMO)
OSX also has features Linux doesn't, such as Applescript.
It sort of depends upon what you are doing with it. I'm not an expert on Active Directory. I did a bit with Novell's equivalent. But I've not done much with Microsoft's AD. I do recall a write up someone did that went through the limitations, fixes, work arounds and so forth. But I believe the general consensus is that Panther is necessary to be a full player.
There are some limitations and difficulties with Active Directory under Jaguar. However I believe that Panther (coming out in a little over a week) is a full Active Directory player and can even store your home directory on the Windows server for proper roaming.
No joke. Check out newsgroups like alt.binaries.smallville, alt.binaries.multimedia.smallville, alt.binaries.multimedia.alias etc.
If you use the newsgrab perl script as a cron job then there really is no hassle. I just check my directory the following day and there is the MPEG of the HDTV feed. (Typically only 1 GB with commercials removed and remember that the stations don't broadcast in anywhere near the highest HDTV resolution)
Most feature bloat doesn't slow a program down. It adds unnecessary functions often. But those are typically accessed only rarely. Background spellchecking certainly will slow things down. But damn if I'd want to be without it. That's one of the nice things of Safari as a web browser -- integrated background spellchecking for editing windows. I can definitely tell the difference between posts here I write at work on Windows and those I write at home on my Mac.
Some programs are slower not because the program is necessarily slower but because you are doing more. Yea WP5.1 was fast. But it didn't have formated text, anti-aliasing and so forth. You also didn't tend to include graphics and you typically weren't multitasking.
Now some programs are admittedly poorly written and not optimized. But that is related to feature bloat only to the degree that programmers time was spent on those other features rather than optimizing the core features. You can see that in many products. For instance OSX 10.3 is much faster than 10.2. Why? Optimization. Same with certain functions in the latest Adobe apps.
It's always a trade off and I agree that feature bloat is a problem. However to me the problem is more in the UI rather than speed.
My wife has freckles, so Jennifer Garner isn't necessary. Further I download all the HDTV feeds of Alias, Smallville, and 24 off USENET. Yeah I wait a day, but it works great. I have a cron task that downloads the episodes I need.
I do have a question though. Comcast is doing HDTV but I believe all the Linux HDTV cards only support over the air broadcasts. Anyone know if there are any cards that word with cable or dish HDTV broadcasts? (For those HDTV Animal Planet shows)
The writeup was a little short on details. Does this work with Dish network accounts? i.e. does it store the MPEG feed of Dish without the D->A->D mess and loss? Exactly how well does it integrate into say Comcast digital cable?
I was intrigued but wanting more information.
Also, I notice Comcast is finally doing HDTV feeds. How do HDTV cards handle that? I assume they don't include software/controls for handling channel switching easily. (I recognize that the box that is the topic of this article doesn't support HDTV)
The only reason to Spam is to sell a product. But surely if some seller advertises this way, utilizing hacked systems, they are in serious violation of law. Why don't the feds simply go after the clients of spammers. If that happened enough you'd think that the spammers wouldn't be able to make money and would simply stop spamming!
Before I start, could I suggest that Slashdot change tags so hitting back and forward doesn't erase the content of ones editing? (grumble, grumble) I've lost text I've been editing while hitting back to see someone else's point far too many times.
Anyway, the above is partially true. It is true that iTunes is a Carbon app. And Carbon isn't available for Windows. However back when Apple ported Quicktime to Windows they did port a lot of the old MacOS toolbox. A lot of that toolbox is in Carbon, albeit in a rewritten way. So the port isn't necessarily as complex to windows as some might suggest.
Of course many people have complained about Quicktime for Windows and a lot of those complaints aren't just about the UI/features of the player. Those complaints probably end up being about the quality of the API that quicktime provides for this pseudo-Carbon-like toolkit.
Now I imagine that the people working on iTunes for Windows and QT for Windows aren't just working on a straight port, but also a better (and more compatible) port of the Carbon libraries for Windows. I don't claim to have any inside information on this. But it seems likely that a lot of the code for iTunes for Windows will be the same as the Mac version. Further porting the underlying library will make other ports easier.
One should point out that with regards to Quicktime, a lot of the development has been to make it better multithreaded and re-entrant. (Or so the word around the campfire goes...) One can't help but suspect that this rewrite would carry over to Windows unless Apple plans on dropping QT for Windows.
Given all this, one suspects that better Windows programs are in the road ahead.
The big issue would be the debate about who invented calculus. I want to know how that would have gone over had it happened today. Huge plagiary debates! Different notations! Huge egos! Different philosophies!
If TIVO had HDTV support and a nice plug into Comcast's cable HDTV then I'd probably use it in a second.
I looked at Dish, but it doesn't appear to host that many HDTV shows yet.
I don't have an HDTV yet, but have been eying them as they come down in price and more and more media is becoming available.
Re:The one i hate most
on
Software Fashion
·
· Score: 2, Insightful
Damn yes. Microsoft had some of the stupidest variable names because of that. What was so bad with just looking up the damn variable's definition! (Rather easy with most modern IDEs) Some of the naming schemes for variables are just ridiculous.
Most movements like this or XP have their points. It's just that the good is almost overwhelmed by the "let's do it just because." People get caught up in the title rather than the content. This is exactly what leads to the dreaded "checklist bloat" of most products. They have to deal with "buzzwords" regardless of function or use.
Re:Let's vote for the greatest forgotten...
on
Software Fashion
·
· Score: 3, Interesting
C# (as opposed to the generic.NET) actually is a good language that does improve on Java a lot. One can even argue that its GUI tools are a considerable improvement over Swing.
On the other hand one must ask whether programming in C# really is better than doing RAD in C++ Builder or even (ugh) Delphi.
Also, while it may be a "fad" at the moment, we should remember that Java was as well for a long time. Yet Java definitely made it past the fad moment. With C# this is even more likely since it seems like Microsoft will be making Windows more and more dependent upon.NET whether we like it or not. Thus calling it a "fad" seems difficult, despite all the exaggerated hype.
You forget that in those days PC users looked down their noses at GUIs. Further Apple didn't support Lotus 1-2-3 or Wordperfect natively. There also was *more* of a software compatibility issue rather than less.
...for example they added color and quicktime, which took years for the shipping version of HyperCard to gain...
This is true and is one of the many reasons Hypercard didn't do as well as it could have. It took too long to add features everyone else was demanding. Color and more robust scripting being but a few of the features. Then Apple basically neglected it instead of doing things like adding better Applescript support. (All of this during the nadir of Apple in the mid-90's)
I wonder if they will have to deal with that "layer change" pause that DVDs have. Don't they have a way of knowing where the data is stored and break up the film accordingly?
Slashdot Mirror
I think you miss the point. SPEC isn't necessarily telling for real world speeds - especially on dual processor units. I agree SPEC is important. But it certainly isn't the whole story.
Ars Discussion of Athalon-64 vs. G5
We'll see what happens. Competition is good for everyone. I prefer, for a variety of reasons, OSX. And by and large I think performance gets too much focus by people who will never use it. But at the same time iMovie and FCE shows digital editing is coming to the masses and something faster than even a G5 or Athalon-64 would be nice there.
But I agree that Mathematica would be a better choice. The problem with the test was that it was confused about whether it was testing platforms to do common tasks or simply comparing the speed of the platforms. For the former then it really should have used better applications on the Mac, such as FCP. For the latter it also chose a rather questionable suite.
Further I think the criticisms of the applications tested is entirely appropriate. It really was a very questionable test. I think the Athalon-64 would win a fair test as well. Just that this wasn't that test.
Anyone who thinks they'll get better Officer performance on either XP or OSX with the new chips are right, but wrong at the same time. A 500 MHz PIII runs OfficeXP handily with no speed perception problems. Anything else, unless you are really hitting complex Excel worksheets (doubtful), is overkill. Same on the Mac, albeit slightly less so due to Microsoft's code. But lets say a 800 MHz G4 to be fully satisfied.
For the vast majority of users faster chips are overkill. They are of benefit to people doing complex graphics, scientific programming, databases, and games. And yes, that latter is what counts to many people. I personally consider it idiocy what people spend to play video games, but that's me. But for probably 80% of the people using computers what came out two years ago was more than powerful enough.
To do benchmarks on such programs seems silly to me.
Anyone buying an Athalon-64 primarily to run the latest incarnation of Doom or Office is simply an idiot. If they happen to run those as well, that's fine.
And, as an other poster pointed out, overall the suite of products used to test the computers makes little sense. I mean I expect the Athalon to be faster. Always did. But if you want to make a meaningful comparison use applications that make some sense.
Having said that few people need a full 64-bit OS and Panther is aimed squarely at consumers. I expect 64 bit pointers to come eventually, but it is probably better to stick with 32 bit pointers and keep better compatibility.
And, let's be honest. Most people running the Athalon-64 will be using it just as a fast Athalon running WindowsXP.
I'd reject that assumption for many applications.
The desktop of OSX is vastly superior to Linux on the Desktop of even products like Ximian. Further there is far more productivity software for OSX than Linux.
Realistically though one can ask if a x86 OSX would have all its software ported. So you do have a point. But I think that in terms of user friendliness OSX beats any Linux distro and saying that Linux will get there "real soon now" is still wishful thinking. (IMO)
OSX also has features Linux doesn't, such as Applescript.
It sort of depends upon what you are doing with it. I'm not an expert on Active Directory. I did a bit with Novell's equivalent. But I've not done much with Microsoft's AD. I do recall a write up someone did that went through the limitations, fixes, work arounds and so forth. But I believe the general consensus is that Panther is necessary to be a full player.
There are some limitations and difficulties with Active Directory under Jaguar. However I believe that Panther (coming out in a little over a week) is a full Active Directory player and can even store your home directory on the Windows server for proper roaming.
No joke. Check out newsgroups like alt.binaries.smallville, alt.binaries.multimedia.smallville, alt.binaries.multimedia.alias etc. If you use the newsgrab perl script as a cron job then there really is no hassle. I just check my directory the following day and there is the MPEG of the HDTV feed. (Typically only 1 GB with commercials removed and remember that the stations don't broadcast in anywhere near the highest HDTV resolution)
Some programs are slower not because the program is necessarily slower but because you are doing more. Yea WP5.1 was fast. But it didn't have formated text, anti-aliasing and so forth. You also didn't tend to include graphics and you typically weren't multitasking.
Now some programs are admittedly poorly written and not optimized. But that is related to feature bloat only to the degree that programmers time was spent on those other features rather than optimizing the core features. You can see that in many products. For instance OSX 10.3 is much faster than 10.2. Why? Optimization. Same with certain functions in the latest Adobe apps.
It's always a trade off and I agree that feature bloat is a problem. However to me the problem is more in the UI rather than speed.
I do have a question though. Comcast is doing HDTV but I believe all the Linux HDTV cards only support over the air broadcasts. Anyone know if there are any cards that word with cable or dish HDTV broadcasts? (For those HDTV Animal Planet shows)
I was intrigued but wanting more information.
Also, I notice Comcast is finally doing HDTV feeds. How do HDTV cards handle that? I assume they don't include software/controls for handling channel switching easily. (I recognize that the box that is the topic of this article doesn't support HDTV)
The only reason to Spam is to sell a product. But surely if some seller advertises this way, utilizing hacked systems, they are in serious violation of law. Why don't the feds simply go after the clients of spammers. If that happened enough you'd think that the spammers wouldn't be able to make money and would simply stop spamming!
Anyway, the above is partially true. It is true that iTunes is a Carbon app. And Carbon isn't available for Windows. However back when Apple ported Quicktime to Windows they did port a lot of the old MacOS toolbox. A lot of that toolbox is in Carbon, albeit in a rewritten way. So the port isn't necessarily as complex to windows as some might suggest.
Of course many people have complained about Quicktime for Windows and a lot of those complaints aren't just about the UI/features of the player. Those complaints probably end up being about the quality of the API that quicktime provides for this pseudo-Carbon-like toolkit.
Now I imagine that the people working on iTunes for Windows and QT for Windows aren't just working on a straight port, but also a better (and more compatible) port of the Carbon libraries for Windows. I don't claim to have any inside information on this. But it seems likely that a lot of the code for iTunes for Windows will be the same as the Mac version. Further porting the underlying library will make other ports easier.
One should point out that with regards to Quicktime, a lot of the development has been to make it better multithreaded and re-entrant. (Or so the word around the campfire goes...) One can't help but suspect that this rewrite would carry over to Windows unless Apple plans on dropping QT for Windows.
Given all this, one suspects that better Windows programs are in the road ahead.
The big issue would be the debate about who invented calculus. I want to know how that would have gone over had it happened today. Huge plagiary debates! Different notations! Huge egos! Different philosophies!
In OSX it is ls | pbcopy
Nice to know gun control is working so well up there...
If TIVO had HDTV support and a nice plug into Comcast's cable HDTV then I'd probably use it in a second. I looked at Dish, but it doesn't appear to host that many HDTV shows yet. I don't have an HDTV yet, but have been eying them as they come down in price and more and more media is becoming available.
Of those 40 Mac viruses how many work under OSX?
Most movements like this or XP have their points. It's just that the good is almost overwhelmed by the "let's do it just because." People get caught up in the title rather than the content. This is exactly what leads to the dreaded "checklist bloat" of most products. They have to deal with "buzzwords" regardless of function or use.
On the other hand one must ask whether programming in C# really is better than doing RAD in C++ Builder or even (ugh) Delphi.
Also, while it may be a "fad" at the moment, we should remember that Java was as well for a long time. Yet Java definitely made it past the fad moment. With C# this is even more likely since it seems like Microsoft will be making Windows more and more dependent upon .NET whether we like it or not. Thus calling it a "fad" seems difficult, despite all the exaggerated hype.
You forget that in those days PC users looked down their noses at GUIs. Further Apple didn't support Lotus 1-2-3 or Wordperfect natively. There also was *more* of a software compatibility issue rather than less.
This is true and is one of the many reasons Hypercard didn't do as well as it could have. It took too long to add features everyone else was demanding. Color and more robust scripting being but a few of the features. Then Apple basically neglected it instead of doing things like adding better Applescript support. (All of this during the nadir of Apple in the mid-90's)
I wonder if they will have to deal with that "layer change" pause that DVDs have. Don't they have a way of knowing where the data is stored and break up the film accordingly?