GCC is the "same" for x86 and G5 only in that they use the same tokenizer. The backends are different, just like icc and gcc are different. Why is this so difficult to grasp?
If SPEC is so "useless", why is Apple making such a big deal of it, with bogus numbers? And as for Photoshop and BLAST, sure, but I bet I could find two programs where a single P4 beats the dual G5 too.
Man, this thread is making me out to be some sort of anti-Mac guy, which I'm not. I just pointed out that Apple is cooking the numbers.
There's no "might have" about it. I gave you the numbers, over 1100 for both specint and specfp. Clearly gcc for x86 generates slower code than Intel's compiler, that's why Apple used it for their benchmarks.
Heck, I even looked up the "rate" numbers for a dual Xeon 3.06 GHz. It's getting 21.7 int and 15.7 fp, compared to the G5's 17.2 and 15.7. With gcc the Xeons only do 16.7 and 11.1, so obviously that's what Apple quotes.
I looked a little bit for results with the MS compilers but didn't find anything, I agree that would be interesting to see.
Man, that post was about the most flamebait-y thing I've ever written, and also the most responded-to. I guess that's how you have to be to get anywhere around here.:-) I mean really, it's obvious that Apple will show off the stuff that makes it look the best, just like anyone else would. It's just that when Apple does it somehow it becomes God's Own Truth.
Of course benchmarks are only approximations of performance, but Apple sure seems to think that SPEC is a good one according to their web page. I'm just pointing out that much better performance has been achieved on the P4 than what they're quoting. Do you dispute my assertion or are you just angry that I pointed it out?
Now, now, don't try to pull me into your Mac-hating club. I actually think OS X is pretty sweet and I like my iPod, and I'm sure the new machines will be quite nice to work with. I just get sick of Apple's BS once in a while and have to let it out. All this gloating from the zealots over a non-existent performance advantage sent me over the edge.
Apple is lying, very egregiously. If you go to veritest.com and download the report, it gives these numbers:
specint_base2000
G5: 800
P4: 889
Xeon: 836
specfp_base2000
G5: 840
P4: 693
Xeon: 646
So, the G5 is the slowest in integer but fastest by quite a bit in FP. But wait: these tests were done using gcc, which nobody in the Intel world would actually use to compile code that needs to run fast. At specbench.org, I see for the P4, using Intel's compiler (hey, gcc is now Apple's compiler, right?)
specint_base2000: 1164
specfp_base2000: 1200
Admittedly, this just proves that gcc sucks, but that's all you get from Apple. Nothing has changed, Intel is still winning and Apple is still lying about it.
Cool, does that mean that Mac-heads will finally admit after all of these years that the G4 systems really weren't a match for x86 boxes after all?
The new machines sound very nice, although I wish Apple would stop pricing them with unusably low amounts of RAM. I'm also a bit skeptical about August delivery dates but we'll see...
My Centrino laptop runs at 1.3 GHz and gets 6 hours of battery life. It also has faster RAM and a larger, higher-resolution screen than your TiBook (15", 1400x1050). Face it, Apple is falling just as behind in the laptop arena as they are in the desktop one.
The whole SCO storyline is unbelievably boring, but I have to admit this sort of impotent rage can be highly amusing. At least we're getting that out of it.
Re:How's about and update for 1st gen Windows IPod
on
New iPod Firmware Released
·
· Score: 2, Informative
You can get a 1.30 updater for Windows here. I installed this a long time ago and it works fine for me.
I guess then they've got 'GVADEC' and '6UADEC' coming up, but what will they do for the seventh one? They'd better get a task force going on that, soon.
At relatively large type sizes (at least 11 pt or so) I agree that OS X is pretty spiffy. However, try going to sfgate.com, for one example. The bulleted subheadlines will look either fuzzy or crooked, depending on whether they get antialiased or not. On XP they look much better to me, although again that could just be the higher-res screen.
I used to have an iBook, and now have a laptop running Windows XP. I'd say that the ClearType fonts on the XP one look far better to me than the ones on the iBook did. Arial in small sizes is especially bad on OS X, if you antialias it it looks too fuzzy but if you don't the hinting makes it look all crooked. Perhaps it's just because this XP machine has a 1400x1050 screen that the fonts look so much better, but on the other hand you can't even get a Mac laptop with that sort of resolution. The end result is all that matters, and I'd never go back to the iBook now.
Where are you getting the numbers (which you didn't bother to quote)? A quick google for the first thing that looked promising turned up this page, which states:
Based on information supplied by IBM, a PPC 970 running at 1.8 GHz is estimated at: SPECfp - 1051 SPECint - 937
Re:Exactly what makes a supplier monopolistic
on
iBox Episode 2
·
· Score: 1
MacOS isn't an industry, personal computers are an industry. Apple does not have a monopoly on personal computers. By your way of thinking any company with patented or otherwise exclusive technology would be considered a monopoly. You may wish that were the case but it's not.
Re:String.Replace("Apple","Microsoft")
on
iBox Episode 2
·
· Score: 1
What criticism is it that I'm rationalizing away? That Apple is a company in business to make money, and they expect their repair shops to live up to their contracts? That should be obvious to anyone and is nothing for them to be ashamed of.
BTW, I don't own any Apple computers. I'm typing this on a laptop running Windows XP and at work I use Linux, so I'm not sure what 'faith' I'm supposed to be defending.
Re:String.Replace("Apple","Microsoft")
on
iBox Episode 2
·
· Score: 2, Funny
Yeah, the reaction would be "Microsoft doesn't make computers".
Re:Exactly what makes a supplier monopolistic
on
iBox Episode 2
·
· Score: 1
What would be monopolistic is if you failed to reveal enough information to someone else to design an alternate computer to run the software on another platform.
What's monopolistic about that? I thought "monopoly" referred to controlling all of a particular market, but I guess on Slashdot it's devolved into a generic replacement for "bad" or "I don't like it".
Re:"Actively searching for new suppliers"?
on
iBox Episode 2
·
· Score: 1
I don't think Apple is getting a free ride because this must be standard policy for nearly all computer manufacturers. Do you think you can just call up Dell or HP and order a bare motherboard? I'm sure they would be happy to sell you one, but only as part of a complete system. Same with Apple.
This reminds me of a story I saw in the Onion, it was called something like "Asian Teens Protest against Stereotyping". The teenagers were complaining that just because they were Asian people thought they were really smart and motivated to succeed, when really they just wanted to lie around all day smoking pot. It was hilarious.
I've seen a bunch of that Cony stuff around here lately, it's hilarious. Their trade dress is a total ripoff of Sony, right down to the graphic design of the boxes, the fonts used, similar model-numbering scheme, etc. Totally shameless, I love it.:-)
Slashdot Mirror
GCC is the "same" for x86 and G5 only in that they use the same tokenizer. The backends are different, just like icc and gcc are different. Why is this so difficult to grasp?
If SPEC is so "useless", why is Apple making such a big deal of it, with bogus numbers? And as for Photoshop and BLAST, sure, but I bet I could find two programs where a single P4 beats the dual G5 too.
Man, this thread is making me out to be some sort of anti-Mac guy, which I'm not. I just pointed out that Apple is cooking the numbers.
There's no "might have" about it. I gave you the numbers, over 1100 for both specint and specfp. Clearly gcc for x86 generates slower code than Intel's compiler, that's why Apple used it for their benchmarks.
Heck, I even looked up the "rate" numbers for a dual Xeon 3.06 GHz. It's getting 21.7 int and 15.7 fp, compared to the G5's 17.2 and 15.7. With gcc the Xeons only do 16.7 and 11.1, so obviously that's what Apple quotes.
I looked a little bit for results with the MS compilers but didn't find anything, I agree that would be interesting to see.
:-) I mean really, it's obvious that Apple will show off the stuff that makes it look the best, just like anyone else would. It's just that when Apple does it somehow it becomes God's Own Truth.
Man, that post was about the most flamebait-y thing I've ever written, and also the most responded-to. I guess that's how you have to be to get anywhere around here.
Of course benchmarks are only approximations of performance, but Apple sure seems to think that SPEC is a good one according to their web page. I'm just pointing out that much better performance has been achieved on the P4 than what they're quoting. Do you dispute my assertion or are you just angry that I pointed it out?
Now, now, don't try to pull me into your Mac-hating club. I actually think OS X is pretty sweet and I like my iPod, and I'm sure the new machines will be quite nice to work with. I just get sick of Apple's BS once in a while and have to let it out. All this gloating from the zealots over a non-existent performance advantage sent me over the edge.
Oops, sorry, the result for the P4 specfp_base2000 should have read 1213 instead of 1200.
Apple is lying, very egregiously. If you go to veritest.com and download the report, it gives these numbers:
specint_base2000
G5: 800
P4: 889
Xeon: 836
specfp_base2000
G5: 840
P4: 693
Xeon: 646
So, the G5 is the slowest in integer but fastest by quite a bit in FP. But wait: these tests were done using gcc, which nobody in the Intel world would actually use to compile code that needs to run fast. At specbench.org, I see for the P4, using Intel's compiler (hey, gcc is now Apple's compiler, right?)
specint_base2000: 1164
specfp_base2000: 1200
Admittedly, this just proves that gcc sucks, but that's all you get from Apple. Nothing has changed, Intel is still winning and Apple is still lying about it.
Cool, does that mean that Mac-heads will finally admit after all of these years that the G4 systems really weren't a match for x86 boxes after all?
The new machines sound very nice, although I wish Apple would stop pricing them with unusably low amounts of RAM. I'm also a bit skeptical about August delivery dates but we'll see...
My Centrino laptop runs at 1.3 GHz and gets 6 hours of battery life. It also has faster RAM and a larger, higher-resolution screen than your TiBook (15", 1400x1050). Face it, Apple is falling just as behind in the laptop arena as they are in the desktop one.
The whole SCO storyline is unbelievably boring, but I have to admit this sort of impotent rage can be highly amusing. At least we're getting that out of it.
You can get a 1.30 updater for Windows here. I installed this a long time ago and it works fine for me.
I guess then they've got 'GVADEC' and '6UADEC' coming up, but what will they do for the seventh one? They'd better get a task force going on that, soon.
It's because it's a GNU project, and GNU is pronouced 'guh-new' according to gnu.org. I don't like it any more than you do, but that's how it is...
At relatively large type sizes (at least 11 pt or so) I agree that OS X is pretty spiffy. However, try going to sfgate.com, for one example. The bulleted subheadlines will look either fuzzy or crooked, depending on whether they get antialiased or not. On XP they look much better to me, although again that could just be the higher-res screen.
Yeah, that's really worth $130 to someone who has a 32-bit machine (i.e. every single Mac owner).
I used to have an iBook, and now have a laptop running Windows XP. I'd say that the ClearType fonts on the XP one look far better to me than the ones on the iBook did. Arial in small sizes is especially bad on OS X, if you antialias it it looks too fuzzy but if you don't the hinting makes it look all crooked. Perhaps it's just because this XP machine has a 1400x1050 screen that the fonts look so much better, but on the other hand you can't even get a Mac laptop with that sort of resolution. The end result is all that matters, and I'd never go back to the iBook now.
Where are you getting the numbers (which you didn't bother to quote)? A quick google for the first thing that looked promising turned up this page, which states:
Based on information supplied by IBM, a PPC 970 running at 1.8 GHz is estimated at:
SPECfp - 1051
SPECint - 937
For comparison, 3.06GHz Pentium4 scores roughly:
SPECfp - 1077
SPECint - 1099
So much for "trashing" the P4...
MacOS isn't an industry, personal computers are an industry. Apple does not have a monopoly on personal computers. By your way of thinking any company with patented or otherwise exclusive technology would be considered a monopoly. You may wish that were the case but it's not.
What criticism is it that I'm rationalizing away? That Apple is a company in business to make money, and they expect their repair shops to live up to their contracts? That should be obvious to anyone and is nothing for them to be ashamed of.
BTW, I don't own any Apple computers. I'm typing this on a laptop running Windows XP and at work I use Linux, so I'm not sure what 'faith' I'm supposed to be defending.
Yeah, the reaction would be "Microsoft doesn't make computers".
What would be monopolistic is if you failed to reveal enough information to someone else to design an alternate computer to run the software on another platform.
What's monopolistic about that? I thought "monopoly" referred to controlling all of a particular market, but I guess on Slashdot it's devolved into a generic replacement for "bad" or "I don't like it".
I don't think Apple is getting a free ride because this must be standard policy for nearly all computer manufacturers. Do you think you can just call up Dell or HP and order a bare motherboard? I'm sure they would be happy to sell you one, but only as part of a complete system. Same with Apple.
This reminds me of a story I saw in the Onion, it was called something like "Asian Teens Protest against Stereotyping". The teenagers were complaining that just because they were Asian people thought they were really smart and motivated to succeed, when really they just wanted to lie around all day smoking pot. It was hilarious.
Oops, I meant Coby, not Cony. They're not THAT shameless...
I've seen a bunch of that Cony stuff around here lately, it's hilarious. Their trade dress is a total ripoff of Sony, right down to the graphic design of the boxes, the fonts used, similar model-numbering scheme, etc. Totally shameless, I love it. :-)