MacCoaster
Oct 10, 02:21 AM
Originally posted by javajedi
Someone inquired about the benchmark Java console program I created:
It's located at http://members.ij.net/javajedi
I've also included the source (FPMathTest.java) for the curious.
<snip>
Kevin
That was me. :)
Thanks. See above post for my results. I even ported your Java code to C# (so similar, it scared me!) and got slightly lower numbers.
8152, 8151, 8162, 8142, 8172, 8142, 8161, 8152... all for a final average of 8154.25.
[edit: whoa, recently got 7891... running it more to average]
[edit #2: 7891, 7892, 7902, 7891, 7882, 7892, 7882, 7881... all for a final average of 7889.125]
You may have the source code/binary to test on your Windows computer (or Linux, with Mono; or BSD, with Microsoft's ROTOR)--just give me a hoot.
Someone inquired about the benchmark Java console program I created:
It's located at http://members.ij.net/javajedi
I've also included the source (FPMathTest.java) for the curious.
<snip>
Kevin
That was me. :)
Thanks. See above post for my results. I even ported your Java code to C# (so similar, it scared me!) and got slightly lower numbers.
8152, 8151, 8162, 8142, 8172, 8142, 8161, 8152... all for a final average of 8154.25.
[edit: whoa, recently got 7891... running it more to average]
[edit #2: 7891, 7892, 7902, 7891, 7882, 7892, 7882, 7881... all for a final average of 7889.125]
You may have the source code/binary to test on your Windows computer (or Linux, with Mono; or BSD, with Microsoft's ROTOR)--just give me a hoot.
R.Perez
Mar 13, 04:00 PM
The biggest wind farm in the world provides around 2MW/km^2. Your 100milesX100miles plant would only provide around 52 000MW (52GW) of power with same ratio. USA's power consumption in 2005 was 29PWh. I don't know how exactly this things can be converted but Fukushima I has installed power of 4.7GW and provides 25.8GWh each year while the biggest wind farm has installed capacity of 781MW. The plant you described would be around 10 times more powerful than the Fukushima but even then, it could provide around 250GWh which is a fraction of 29PWh.
If someone knows how to convert these things properly or has more info on this, please educate me/us.
Did I say at any point time that we should rely on just wind? or solar, or tidal for that matter? A combination of all three is in order here. On top of that re-thinking infrastructure so that at least some of the power can be generated from the home or building itself is in order. i.e. putting solar panels on all new construction. This would reduce the amount of energy needed from centralized sources. Also shifting towards smarter energy consumption would help as well, i.e. using geo-thermal to generate heat instead of oil or electricity and mandating more efficient lightbulbs and appliances.
If someone knows how to convert these things properly or has more info on this, please educate me/us.
Did I say at any point time that we should rely on just wind? or solar, or tidal for that matter? A combination of all three is in order here. On top of that re-thinking infrastructure so that at least some of the power can be generated from the home or building itself is in order. i.e. putting solar panels on all new construction. This would reduce the amount of energy needed from centralized sources. Also shifting towards smarter energy consumption would help as well, i.e. using geo-thermal to generate heat instead of oil or electricity and mandating more efficient lightbulbs and appliances.
more...
leekohler
Mar 26, 01:04 AM
Prove why I should be denied the right to copulate in public, and think of the children is not an acceptable answer
WTF? Who said that anyone should be copulating in public? You have completely lost this argument at this point. Not to mention your mind...This has just gotten stupid.
WTF? Who said that anyone should be copulating in public? You have completely lost this argument at this point. Not to mention your mind...This has just gotten stupid.
more...
linknprk
Mar 18, 02:52 AM
So if you're sticking at 4.1.0 and they aren't monitoring, then they should be monitoring 3.x even less, no?
All the more reason for me to stick with 3.1.3 on my 3G.
BL.
um... did you guys misread the article?
The article is proposing that they might be able to suspect unsupported tethering for people NOT using 4.3 because hotspot wasn't made available until 4.3
So if you stick with 4.1 or 3.1.3 or anything earlier than 4.3 (while using data in a way that looks like tethering)... you will stand out.
Thats how I interpreted the article.
All the more reason for me to stick with 3.1.3 on my 3G.
BL.
um... did you guys misread the article?
The article is proposing that they might be able to suspect unsupported tethering for people NOT using 4.3 because hotspot wasn't made available until 4.3
So if you stick with 4.1 or 3.1.3 or anything earlier than 4.3 (while using data in a way that looks like tethering)... you will stand out.
Thats how I interpreted the article.
more...
dobbin
Sep 20, 03:45 AM
I don't want to have to put yet another box on the shelf under my TV and have yet another remote control kicking around my living room.
I already have a DVD, a VCR, and a Sky+ box (DVR). I know that in theory I should choose just one or two of these, but that doesn't work in practice. What happens when my mum records something on a video for me - I still need a VCR, and until Sky bring out a Sky+ box with a *much* larger hard disk then I'll need a DVD for keeping things long term.
If Apple could include at least a DVD burner and ideally a DVR hard disk as well, then I could actually start replacing the other machines I have rather than just adding to them and cluttering up my living room.
Its probably a moot point anyway as I doubt iTV will be launched in the UK for a long while anyway.
I already have a DVD, a VCR, and a Sky+ box (DVR). I know that in theory I should choose just one or two of these, but that doesn't work in practice. What happens when my mum records something on a video for me - I still need a VCR, and until Sky bring out a Sky+ box with a *much* larger hard disk then I'll need a DVD for keeping things long term.
If Apple could include at least a DVD burner and ideally a DVR hard disk as well, then I could actually start replacing the other machines I have rather than just adding to them and cluttering up my living room.
Its probably a moot point anyway as I doubt iTV will be launched in the UK for a long while anyway.
Doctor Q
Mar 19, 12:31 PM
Seriously: if I walk in to a store and take CD from the shelf, and not pay it, I'm stealing. If I make an identical copy of the CD and leave the original on the shelf, I'm not stealing, I'm committing a copyright-infringment. But I'm not stealing.We've had this dictionary discussion before. But when a book author finds somebody using a photocopier to make a copy of their book instead of buying it, the word used doesn't matter as much as the fact you got something they were selling without paying.Same logic: if I take someone else's car, and drive away with it, I'm stealing it. But if I create an identical copy of the car (using a replicator I got from Star Trek) for myself, have I stolen anything? From whom have I stolen?Same logic: Musical artists aren't selling you round bits of plastic. They are selling you a copy of their music. Same logic: When you buy PhotoShop, you are buying more than the CD and some packaging. You are buying a license to use it, and even if you download a copy without taking something away from somebody else, you are getting something worth money and the owner/producer has reason to expect payment.I find it rather surprising how blindly people here defend Apple, even after seeing how they remove your rights little by little. How many times can you burn your iTunes-songs to CD? It used to be ten times. But Apple reduced it to seven.Yeah, and I wonder why they did that. It was at the same time they increased the number of Macs you can authorize, so overall it was an improvement. Maybe they were tinkering with their deal with the record labels.Then they removed the ability to share/stream your songs from itunes to others.I can't imagine how they made that mistake, allowing sharing over the Internet instead of only over LANs when anybody could tell you the record labels (yes, them again) would be up in arms.Little by little, you feel the DRM-noose tightening around your necks. It seems like a major PR-coup to me, when you have Apple reducing your rights little by little, and you guys are screaming "Yes! Reduce our rights even more!"I wonder if they could offer a new program: You get to have all DRM removed in exchange for burly RIAA enforcers paying you surprise visits whenever they like to check what you are listening to. I'm just kidding, but it's too bad that honest customers have to bear the burdens of dishonest customers, and that any of us have to feel hogtied.
more...
Bill McEnaney
Mar 27, 06:16 PM
Everyone, as usual I'm answering posts in a non-chronological order. I'm not ignoring anyone. I need to think hard about what to write about a post by Gelfin. So I may need two or three days to think about it.
Eraserhead wants peer-reviewed scientific articles, so I'll look for them, too. I already have an article in mind by a secular author named "Spitzer" who helped the American Psychiatric Association normalize homosexuality before he changed his mind about that normalization.
Meanwhile, please listen to Nicolosi's first answer in video 3 of the first set of videos, the last part of the three-part interview, where he says that homosexuals have a right to live a gay lifestyle (http://www.josephnicolosi.com/videos2/). That doesn't sound like what a brainwasher would say, does it?
Eraserhead wants peer-reviewed scientific articles, so I'll look for them, too. I already have an article in mind by a secular author named "Spitzer" who helped the American Psychiatric Association normalize homosexuality before he changed his mind about that normalization.
Meanwhile, please listen to Nicolosi's first answer in video 3 of the first set of videos, the last part of the three-part interview, where he says that homosexuals have a right to live a gay lifestyle (http://www.josephnicolosi.com/videos2/). That doesn't sound like what a brainwasher would say, does it?
more...
javajedi
Oct 9, 04:34 PM
Originally posted by TheFink
One point you are missing is that I can upgrade my PC 5 times over and still have the cost be lower than buying a new Mac. So a mac can run modern apps 5 years later. For the same price, I can get a PC, drop a new HD, video card, and CPU in a few years later and then end up with a leading edge PC, and not a bleading edge mac. My B&W G3 isn't even upgradable to the speeds of the current iMacs. With a PC a new mobo and CPU will get me into whatever is the current CPU class....
Very ture. For better or wose, that is what happens when you get locked into a single vendor that sells proprietary hardware *or* software.... just look at Sun :)
One point you are missing is that I can upgrade my PC 5 times over and still have the cost be lower than buying a new Mac. So a mac can run modern apps 5 years later. For the same price, I can get a PC, drop a new HD, video card, and CPU in a few years later and then end up with a leading edge PC, and not a bleading edge mac. My B&W G3 isn't even upgradable to the speeds of the current iMacs. With a PC a new mobo and CPU will get me into whatever is the current CPU class....
Very ture. For better or wose, that is what happens when you get locked into a single vendor that sells proprietary hardware *or* software.... just look at Sun :)
Silentwave
Jul 11, 11:19 PM
Yeah, I hope apple lower their price point for the pro models. It is way too much. I love mac computer, but come on; the prices vs the PC suckass.
I know Macs are way better then PC, but PCs are good tool too.
As I said above, I don't think the difference will be terribly huge. But if apple doesn't move the prices of their top computers too much we'll be in for a good price comparitively. Apple is also likely to offer a larger amount of RAM available. The Dell workstations I configured in the post could only be configured with up to 4GB RAM, anything else you had to add yourself. Of course, apple RAM may be more expensive by a good deal so we will probably see a lot of people buying 3rd party here.
And lets see if apple has another quad at 2x 2.3ghz dual core, that saves $520 more versus the 2x2.66 DC quad, or $1340 versus the 2x3GHz DC quad. Past that point you have diminishing returns. Quad core 2x2.0GHz dual core saves you $260 over the 2.3GHz, $780 over the 2.66, or $1600 over the 3GHz.
I know Macs are way better then PC, but PCs are good tool too.
As I said above, I don't think the difference will be terribly huge. But if apple doesn't move the prices of their top computers too much we'll be in for a good price comparitively. Apple is also likely to offer a larger amount of RAM available. The Dell workstations I configured in the post could only be configured with up to 4GB RAM, anything else you had to add yourself. Of course, apple RAM may be more expensive by a good deal so we will probably see a lot of people buying 3rd party here.
And lets see if apple has another quad at 2x 2.3ghz dual core, that saves $520 more versus the 2x2.66 DC quad, or $1340 versus the 2x3GHz DC quad. Past that point you have diminishing returns. Quad core 2x2.0GHz dual core saves you $260 over the 2.3GHz, $780 over the 2.66, or $1600 over the 3GHz.
more...
Liquorpuki
Mar 14, 06:04 PM
It would require a multi-tiered approach. We have abundant coal which I believe can be made to burn cleanly although I'm not necessarily advocating that. And none of these sources if they break down (except nuclear) threaten huge geographical areas with basically permanent radioactivity. In case of worst case accidents, it could be plowed under but we'd still have substantial problems. The thing about nuclear power if it was perfect it would be a great power source, but it is far from perfect and the most dangerous.
I think technology has to advance first and we got no choice but to be patient.
- Clean coal and coal capture tech needs to advance so we know what to do with the extracted carbon. Right now it just sits underground, like nuclear waste and we hope it doesn't leak.
- Nuclear waste treatment tech needs to advance so the decay rate of waste can be accelerated, and the amount of waste reduced
- Grid energy storage tech needs to advance so renewables can be integrated into base load and we can phase out fossil fuels and nuclear.
- Smartgrid tech needs to get integrated. For everyone worried about electricity consumption, Smartgrid is an answer.
And the things that are going on in Japan are a result of stupid engineering. The engineers did not design the plant to adequately protect the backup generator switchgear. If they did, there'd be no danger of a meltdown right now. And if what Goto said is correct, there's also an engineering ethics issue involved with the containment vessel not being designed to an appropriate standard. This is an example of why engineering is so damn important. Even with an earthquake or tsunami, this was totally preventable.
The fact remains that most of America's energy problems are caused by conspicuous consumption.
If you're talking about energy consumption, yeah, and that's primarily because of oil. If you're talking about electricity consumption, we're actually not that bad.
And the solution is to shift reliance on oil to reliance on electricity. Which is why electric vehicles are gonna be big in the future.
I think technology has to advance first and we got no choice but to be patient.
- Clean coal and coal capture tech needs to advance so we know what to do with the extracted carbon. Right now it just sits underground, like nuclear waste and we hope it doesn't leak.
- Nuclear waste treatment tech needs to advance so the decay rate of waste can be accelerated, and the amount of waste reduced
- Grid energy storage tech needs to advance so renewables can be integrated into base load and we can phase out fossil fuels and nuclear.
- Smartgrid tech needs to get integrated. For everyone worried about electricity consumption, Smartgrid is an answer.
And the things that are going on in Japan are a result of stupid engineering. The engineers did not design the plant to adequately protect the backup generator switchgear. If they did, there'd be no danger of a meltdown right now. And if what Goto said is correct, there's also an engineering ethics issue involved with the containment vessel not being designed to an appropriate standard. This is an example of why engineering is so damn important. Even with an earthquake or tsunami, this was totally preventable.
The fact remains that most of America's energy problems are caused by conspicuous consumption.
If you're talking about energy consumption, yeah, and that's primarily because of oil. If you're talking about electricity consumption, we're actually not that bad.
And the solution is to shift reliance on oil to reliance on electricity. Which is why electric vehicles are gonna be big in the future.
more...
ddtlm
Oct 7, 03:53 PM
Backtothemac:
Jesus you still don't get it. If you compare Apples to Apples, the 1.6GHZ Dual Athlon is still slower in apps that are multi processor aware. Now, how about the PIV? How does that stack up? The x86 is garbage. Any real IT director would know that.
No, I "get it" fine. Don't bother testing a 1.6ghz dual Athlon when 1.8ghz dual Athlons are readily available. It would do you good to note that this test did not cover all "apps that are multi processor aware", it covered only two apps that are multi-processor aware, and on one of them the Mac looses by a lot. Even on its one win, the dual 1.25 G4 would still loose to a top-of-the-line dual Athlon. Which is slower than a top-of-the-line dual Xeon. Get it?
Jesus you still don't get it. If you compare Apples to Apples, the 1.6GHZ Dual Athlon is still slower in apps that are multi processor aware. Now, how about the PIV? How does that stack up? The x86 is garbage. Any real IT director would know that.
No, I "get it" fine. Don't bother testing a 1.6ghz dual Athlon when 1.8ghz dual Athlons are readily available. It would do you good to note that this test did not cover all "apps that are multi processor aware", it covered only two apps that are multi-processor aware, and on one of them the Mac looses by a lot. Even on its one win, the dual 1.25 G4 would still loose to a top-of-the-line dual Athlon. Which is slower than a top-of-the-line dual Xeon. Get it?
KnightWRX
May 2, 09:05 AM
WOW! Malware that requires the user to do a Google search, then download, and install. For all of this, it asks for your credit card number.
How can we ever defend our computers against such a diabolical threat?!
Hum, download and install are automatic. Good thing I don't use Safari.
As I understand it, Safari will open the zip file since it's a "safe" download. But that doesn't mean it'll execute the code within that zip file, so how is this malware executing without user permission?
I haven't seen this malware first hand, but a zip file can be made with absolute paths, making "unzipping" the file put everything where it needs to be to start up automatically on next log in/reboot.
Who's the brainiac who made zip files "safe" ?
so much for the no malware on macs myth :D
funny how the apple fanboys are getting all defensive :rolleyes:
No viruses on the Mac. There's been malware for OS X for quite a while now.
How can we ever defend our computers against such a diabolical threat?!
Hum, download and install are automatic. Good thing I don't use Safari.
As I understand it, Safari will open the zip file since it's a "safe" download. But that doesn't mean it'll execute the code within that zip file, so how is this malware executing without user permission?
I haven't seen this malware first hand, but a zip file can be made with absolute paths, making "unzipping" the file put everything where it needs to be to start up automatically on next log in/reboot.
Who's the brainiac who made zip files "safe" ?
so much for the no malware on macs myth :D
funny how the apple fanboys are getting all defensive :rolleyes:
No viruses on the Mac. There's been malware for OS X for quite a while now.
w_parietti22
Jul 11, 10:19 PM
Xeon! Conroe (Core 2 Duo)is going in the iMac
more...
shawnce
Jul 12, 03:16 PM
I believe the max TDP of G5 processor is something like 80 W.
The below lists power consumed by the part, they are not TDP numbers (only part of the power consumed by a chip leaves the chip as heat, heat is what you have to dissipate and is what TDP attempts define).
PPC 970fx power optimized part (@ 2GHz)
40W average, 45-50 W max, 23 W throttle back (half frequency)
PPC 970fx standard part (@ 2GHz)
48W average, 55-60 W max, 29 W throttle back (half frequency)
To me this puts the PPC 970fx below the TDP of a Conroe... I would say the TDP for the PPC 970fx (@2Ghz) is around 40 W (if not lower).
The below lists power consumed by the part, they are not TDP numbers (only part of the power consumed by a chip leaves the chip as heat, heat is what you have to dissipate and is what TDP attempts define).
PPC 970fx power optimized part (@ 2GHz)
40W average, 45-50 W max, 23 W throttle back (half frequency)
PPC 970fx standard part (@ 2GHz)
48W average, 55-60 W max, 29 W throttle back (half frequency)
To me this puts the PPC 970fx below the TDP of a Conroe... I would say the TDP for the PPC 970fx (@2Ghz) is around 40 W (if not lower).
more...
MacRumors
Mar 18, 02:22 PM
According to Corante.com (http://www.corante.com/copyfight/archives/2005/03/17/johansen_creates_drmfree_interface_to_itunes.php), from the same authors of QTFairUse (http://www.macrumors.com/pages/2003/11/20031122001549.shtml), comes what is described as "the fair interface to the iTunes Music Store". The application called PyMusique (http://fuware.nanocrew.net/pymusique/) acts as a front end for the iTunes Music Store and allows users to preview iTunes songs, signup for an account, buy songs and redownload songs that were bought with PyMusique.
The most notable twist is this quote from Jon Johansen ("DVD Jon"), one of the authors of the application:
It is somewhat interesting from a DMCA/EUCD perspective. The iTunes Music Store actually sells songs without DRM. While iTunes adds DRM to your purchases, PyMusique does not.
Note: This application has been untested by this site, and Apple will likely take steps to prevent future usage.
The most notable twist is this quote from Jon Johansen ("DVD Jon"), one of the authors of the application:
It is somewhat interesting from a DMCA/EUCD perspective. The iTunes Music Store actually sells songs without DRM. While iTunes adds DRM to your purchases, PyMusique does not.
Note: This application has been untested by this site, and Apple will likely take steps to prevent future usage.
more...
ct2k7
Apr 24, 06:29 PM
they left out the interesting parts to keep people guessing
It just flows... it just works..
It just flows... it just works..
jefhatfield
Oct 10, 11:45 AM
Originally posted by benixau
Dear lord,
If you have any heart for 25 million of your wiser men, please make apple use the power4 chip at lightning speeds, and please lord, do it soon. It is becoming hard for us mac men to defend ourselves.
PS. If you could give me a brand new top of the line mac while your at it i wouldnt mind either.
Edit: There is no blasphemy intended here
it reminds me of that janis joplin song...imagine the music in the background ;)
when motorola was stuck at 500 mhz for 18 months, then i started becoming very vocal about dropping them on the high end stuff and going to ibm when there were rumors that they can make a faster chip
but motorola has climbed, though somewhat slowly, out of their pit, and are doing ok
sure, it may be many months ahead when the pcs hit 3.5 ghz+, but i think motorola will deliver a 2 ghz chip sometime next year
and really, after 2 ghz, does anyone think speed will be a top five factor in why one buys a machine?
when i was a computer salesman in 1999, speed and price were the top two issues...and while price will remain a top issue until all computers get really cheap, i think speed will diminish in its importance for the average consumer
many seem to like the dual 1.25 ghz machines and we know apple will speed bump their whole line of powermacs early next year, if not sooner
a lot of the complaints about apple's speed on their machines sound like a lot of benchmark stats reading and not real world usage
most of us do email, internet, word processing, spreadsheets, and light graphics most of the time in the majority of users in the field...of the many machines i sold, i did not come across one high end graphics user or audio professional who needed more than our store was able to provide them
Dear lord,
If you have any heart for 25 million of your wiser men, please make apple use the power4 chip at lightning speeds, and please lord, do it soon. It is becoming hard for us mac men to defend ourselves.
PS. If you could give me a brand new top of the line mac while your at it i wouldnt mind either.
Edit: There is no blasphemy intended here
it reminds me of that janis joplin song...imagine the music in the background ;)
when motorola was stuck at 500 mhz for 18 months, then i started becoming very vocal about dropping them on the high end stuff and going to ibm when there were rumors that they can make a faster chip
but motorola has climbed, though somewhat slowly, out of their pit, and are doing ok
sure, it may be many months ahead when the pcs hit 3.5 ghz+, but i think motorola will deliver a 2 ghz chip sometime next year
and really, after 2 ghz, does anyone think speed will be a top five factor in why one buys a machine?
when i was a computer salesman in 1999, speed and price were the top two issues...and while price will remain a top issue until all computers get really cheap, i think speed will diminish in its importance for the average consumer
many seem to like the dual 1.25 ghz machines and we know apple will speed bump their whole line of powermacs early next year, if not sooner
a lot of the complaints about apple's speed on their machines sound like a lot of benchmark stats reading and not real world usage
most of us do email, internet, word processing, spreadsheets, and light graphics most of the time in the majority of users in the field...of the many machines i sold, i did not come across one high end graphics user or audio professional who needed more than our store was able to provide them
Rt&Dzine
Apr 26, 05:50 PM
Or it vanished in a miracle.
For the bread has risen.
That is too ******* funny!
For the bread has risen.
That is too ******* funny!
Applespider
Mar 20, 06:29 PM
Furthermore, if you are using iTunes music, and you are using iMovie/iDVD, you CAN use tracks in your videos. They import in and you can use them freely in your projects.
Except there have been threads where people did this and when they sent it to friends to view, their computer had to be authorised to do so.
Except there have been threads where people did this and when they sent it to friends to view, their computer had to be authorised to do so.
more...
javajedi
Oct 11, 06:30 PM
Originally posted by javajedi
What you are saying makes a lot of sense. Now that I think about, I too recall reading this somewhere.
Now that we know the real truth about the "better standard FPU", I thought it was time to shed some light on non vectorized G4 integer processing.
It still does 200,000,000 calculations, but this time I'm multiplying ints.
Motorola 7455 G4@800Mhz: 9 seconds (Native)
IBM 750FX G3@700Mhz: 7 seconds (Native)
Intel P4@2600Mhz 2 seconds (Java)
PowerPC 7455 integer processing is consierabley better than floating point (obviously less work doing ints), but still less per cycle than the Pentium 4.
Very intresting the G4 looses both floating point and integer to the IBM chip, at a 100MHz clock disadvantage.
I'm still waiting to see that "better standard FPU" in the G4. It seems the G4 is absolutely useless unless you are fortunate to have vectorized (AltiVec) code.
Alex, yeah, the native version was compiled under 3.1. It really is interesting to note that despite the 750FX's 100MHz clock disadvantage, it is able to outperform it by 22%. Since there is a 13% difference in clock speed, and if clocks were equal, the 750FX is technically 25% more efficient in scalar integer. I should also re-emphasize that I never bothered compiling the test natively for x86, I left it java, so it's not out of the question the P4 could do this in 1 second - and that is *NOT* using any vector libraries, just plain old integer math.
I've found some documentation on the Altivec C programming interface, and this weekend I'm going to make a first attempt at vectorizing it. The integer test should be no problem, but my FPMathTest app that did square roots will be more difficult. With Altivec, there is not recognized double precision floating point, so this complicates doing square roots. If you want more accurate, precision square roots, you have to do Newton Raphson refinement. In other words more ************ you have to go through. I believe in SSE2 you have double precision floating point ops, and if you were to vectorize it, you wouldn't have to compensate for this.
Another theory as to why the P4 is scoring so good is because if I'm not mistaking (and I'm not), the P4's ALU runs at double its clock. So in my case, 5.6GHz. I'm sure this relates to the issue.
I don't know how true this is, but I wouldn't be suprised if there is some truth to it, surely some food for thought:
http://www.osopinion.com/perl/story/17368.html
The G4 was just a hacked-up G3 with AltiVec and an FPU (floating point unit) borrowed from the outdated 604
If this is the case, then no wonder why we are getting these abysmal scores, and no wonder why a 400mhz Celeron can nearly equal it, and no wonder why the 750FX can outperform it (different company, different fpu)
What you are saying makes a lot of sense. Now that I think about, I too recall reading this somewhere.
Now that we know the real truth about the "better standard FPU", I thought it was time to shed some light on non vectorized G4 integer processing.
It still does 200,000,000 calculations, but this time I'm multiplying ints.
Motorola 7455 G4@800Mhz: 9 seconds (Native)
IBM 750FX G3@700Mhz: 7 seconds (Native)
Intel P4@2600Mhz 2 seconds (Java)
PowerPC 7455 integer processing is consierabley better than floating point (obviously less work doing ints), but still less per cycle than the Pentium 4.
Very intresting the G4 looses both floating point and integer to the IBM chip, at a 100MHz clock disadvantage.
I'm still waiting to see that "better standard FPU" in the G4. It seems the G4 is absolutely useless unless you are fortunate to have vectorized (AltiVec) code.
Alex, yeah, the native version was compiled under 3.1. It really is interesting to note that despite the 750FX's 100MHz clock disadvantage, it is able to outperform it by 22%. Since there is a 13% difference in clock speed, and if clocks were equal, the 750FX is technically 25% more efficient in scalar integer. I should also re-emphasize that I never bothered compiling the test natively for x86, I left it java, so it's not out of the question the P4 could do this in 1 second - and that is *NOT* using any vector libraries, just plain old integer math.
I've found some documentation on the Altivec C programming interface, and this weekend I'm going to make a first attempt at vectorizing it. The integer test should be no problem, but my FPMathTest app that did square roots will be more difficult. With Altivec, there is not recognized double precision floating point, so this complicates doing square roots. If you want more accurate, precision square roots, you have to do Newton Raphson refinement. In other words more ************ you have to go through. I believe in SSE2 you have double precision floating point ops, and if you were to vectorize it, you wouldn't have to compensate for this.
Another theory as to why the P4 is scoring so good is because if I'm not mistaking (and I'm not), the P4's ALU runs at double its clock. So in my case, 5.6GHz. I'm sure this relates to the issue.
I don't know how true this is, but I wouldn't be suprised if there is some truth to it, surely some food for thought:
http://www.osopinion.com/perl/story/17368.html
The G4 was just a hacked-up G3 with AltiVec and an FPU (floating point unit) borrowed from the outdated 604
If this is the case, then no wonder why we are getting these abysmal scores, and no wonder why a 400mhz Celeron can nearly equal it, and no wonder why the 750FX can outperform it (different company, different fpu)
izzle22
Sep 21, 01:33 PM
What are you a comedian? Give me a break. I expected this sort of reaction. It's very easy to say that when you're not the one being effected by this.
Hey at least you guys had U2 before we did.:)
Hey at least you guys had U2 before we did.:)
Chundles
Apr 8, 10:22 PM
Great news. Bring on more Infinity Blade-esque games! :D
Hmmm, swipe, swipe, swipe, next. Swipe, swipe, swipe, next.
Nah, gimme the Infinity Blade graphics but in a game that needs more than just flicking left or right.
Hmmm, swipe, swipe, swipe, next. Swipe, swipe, swipe, next.
Nah, gimme the Infinity Blade graphics but in a game that needs more than just flicking left or right.
Photics
Apr 9, 09:47 AM
You know how stereoscopic vision works, right?
I know how the 3DS works, but it was still fun to try. :D
Basically, the 3DS has an 800 x 240 display. It's using double the pixels to recreate the 3D effect, by creating the same image twice, but slightly adjusted to mimic three dimensions.
I think this is a horrible design choice, as the graphics looked blocky to me. I think Apple made the better decision. The extra resolution on the Retina Displays is used to make the graphics more crisp. I think it looks great! A sharper screen makes it more pleasant to use my iPhone, where the 3D effect made it more uncomfortable to use the 3DS.
Before you point out the mote in our eyes, remove the plank from your own.
If I had a plank in my eye, the 3D effect on the 3DS would be useless anyway. :p
I know how the 3DS works, but it was still fun to try. :D
Basically, the 3DS has an 800 x 240 display. It's using double the pixels to recreate the 3D effect, by creating the same image twice, but slightly adjusted to mimic three dimensions.
I think this is a horrible design choice, as the graphics looked blocky to me. I think Apple made the better decision. The extra resolution on the Retina Displays is used to make the graphics more crisp. I think it looks great! A sharper screen makes it more pleasant to use my iPhone, where the 3D effect made it more uncomfortable to use the 3DS.
Before you point out the mote in our eyes, remove the plank from your own.
If I had a plank in my eye, the 3D effect on the 3DS would be useless anyway. :p
Peterkro
Mar 14, 11:06 AM
EDIT: Here's a FANTASTIC read on Fukushima: http://reindeerflotilla.wordpress.com/2011/03/13/all-right-its-time-to-stop-the-fukushima-hysteria/
Yes that is a good article although pro-nuclear.I originally was flummoxed by the bit about bringing in portable generators and not being able to use them because the connecting plugs were different,this apparently is not the case it's that the switchgear is in a room that is flooded with radioactive water and they can't get rid of the water.I've quoted this guy before and whether he has an axe to grind or not he is not as confident in the plant as others seem to be:
"Japanese engineer Masashi Goto, who helped design the containment vessel for Fukushima's reactor core, says the design was not enough to withstand earthquakes or tsunamis and the plant's builders, Toshiba, knew this."
Here's another article from the NYT which may be useful:
http://www.nytimes.com/2011/03/14/world/asia/japan-fukushima-nuclear-reactor.html?pagewanted=1&_r=2&hp
I think it's to early to make any judgements about what's happening.
Yes that is a good article although pro-nuclear.I originally was flummoxed by the bit about bringing in portable generators and not being able to use them because the connecting plugs were different,this apparently is not the case it's that the switchgear is in a room that is flooded with radioactive water and they can't get rid of the water.I've quoted this guy before and whether he has an axe to grind or not he is not as confident in the plant as others seem to be:
"Japanese engineer Masashi Goto, who helped design the containment vessel for Fukushima's reactor core, says the design was not enough to withstand earthquakes or tsunamis and the plant's builders, Toshiba, knew this."
Here's another article from the NYT which may be useful:
http://www.nytimes.com/2011/03/14/world/asia/japan-fukushima-nuclear-reactor.html?pagewanted=1&_r=2&hp
I think it's to early to make any judgements about what's happening.
more...
Комментариев нет:
Отправить комментарий