Phone games are not exactly built to run on a specific hardware
Absolutely correct. No developer in their right mind would use any of the NEON vector instructions available on ARM CPUs. I mean come on, who could possibly want to process 4 floats in the time it takes to process 1? Especially when that code works on all of the ARM android phones too. I mean, what is the point in bothering to write scalable efficient code for a platform with a finite battery life? I can see absolutely no point to it at all. As a game developer myself, I can say with some authority that the entire iGaming industry has adopted the approach of running an SQL server on your iDevice, and then we simply fire an SQL query for every pixel when we come to rendering. It sure as hell beats using that crappy parallel computing language GLSL to render pixels....
Yes....
Admittedly it's not fully featured, however it's also worth bearing in mind that the iPhone/iPad are more powerful devices than the SGI machines I used in the late 90's to create CG animation. It is a matter of time until your phone is 'good enough' to do everything you need. For a large section of the population, I imagine that time has already arrived. For the rest of us, we'll just have to wait another 5 years or so...
One of the biggest myths in the industry, and one that is not exactly true. If you specifically select "Optimise for Intel i7", then guess what, it only runs on an i7. If however you specify the option to optimise for all platforms (/QxO) you'll have no such problems with AMD chips. The Intel compiler can be instructed to insert two codepaths into an exe. So you can have one specified as/QxO, and a separate codepath for i7. At startup, the app will indeed check for the intel vendor string, and will check to see if it's an i7. If it doesn't find it, it will run the codepath optimised for all processors, rather than the one specifically for the i7. In this regard there is absolutely no difference to running the exe on a Core2, or an AMD chip. In both cases the exe will run the second codepath. The problem only arises if the developer has simply chosen one codepath (or failed to select an optimised codepath for the second one, which could indeed look like a deliberate retardation of performance to a layman).
So the complaint about the CPU dispatcher basically boils down to this: Why does specifying "Optimise for Intel i7" mean exactly that, rather than "Optimise for any CPU supporting the same features as an i7". I can't answer that, and I'd certainly prefer the latter, however it is information that is clearly documented.
To be fair, you have a valid point about rigged benchmarks, but your reasoning is entirely incorrect. The simple fact is that the ATOM is significantly different enough from other x86 processors as to make any comparative benchmarking entirely redundant. You either compile a binary explicitly for ATOM (and it runs quite well), or you choose a general x86 compiler option (and it runs like a dog). Most benchmarking software will not have been compiled with the ATOM in mind, so chances are it will run like a dog, making the processor look worse than it actually is....
For anything that uses double-precision floating point the ARM core is very weak, although most of those things can be run on the DSP or GPU (and use even less power).
The ATOM core is extremely weak for double precision (a couple of orders of magnitude slower than single precision). I'm also quite surprised to hear that you've managed to get the GPU to support double precision? Which GPU are you using? The GL_DOUBLE is not a valid argument for glVertexAttribPointer in GLES, so I'd be interested in hearing how you subverted this? I've not tried windows 8 for tablets, so is this something microsoft have added to their D3D equivalent for tablets?
You are talking about the wrong department. This is G.C.H.Q. we are talking about, not Mi5 / Mi6. They are effectively nothing more than the poorly paid secretary to Mi5. Good programming graduates would go on to have interesting and fulfilling career in the software industry. Those that fail to do that, become teachers. Those that fail to do that, go to G.C.H.Q......
You seem to be assuming that people would want to work at GCHQ? I grew up next door to the place, and as someone who was training to be a programmer, GCHQ was that added bit of motivation to do well at university. It was always a case of "work hard, get a good degree, otherwise you'll have to apply to GCHQ.....". I certainly never got the impression it was an inspiring place to work.....
I've heard those arguments said about the British when they first introduced recycling. Strangely enough, everyone recycles their waste just fine now. Maybe it was the education, maybe it was the councils refusal to collect mixed rubbish, or maybe it was the fines they started dishing out to people who constantly refused to recycle. Whatever it was, recycling and composting are second nature to the vast majority of the population now.... I don't see any reason why America should be any different.
I pay my city (I'm in the UK) to collect my recycling, and take away the very small quantity of stuff that's can't be recycled. The ONLY pre-requisite is that organic matter is not mixed in with the recycling. There is absolutely no need to sort the paper from glass.....
The problem of plastics is simply a case of "does it have the recycling symbol on it, or not". Even that seems to be moot these days (I'm assuming someone passed a law to say that all plastics containers must be able to be recycled, because I haven't seen a non-recyclable bit of plastic for years).
To be honest though, even with the councils previous "sort everything" system, I never found it to be too hard. Just collect your recycling in a single box, and sort it when you put it into the bins. It takes what, 2 minutes at most? Not exactly a high price to pay to keep it from being buried in the countryside....
Nothing. The life of the battery will stay the same, it'll just get bigger. Now it's just the job of marketing to persuade the consumer that retro styled 80's bricks are back in fashion.....
Will anyone ever see the performance advantage of a 4-core ARM chip outside of benchmarks?
I must admit I was very very sceptical of the iPad. I acquired a first gen one from the office fanboy, and I might have been somewhat converted to the platform. In almost all situations, I will admit that my cheapio netbook is much more practical tool (I've actually come to the conclusion that anyone using an iPad for 'business' is actually using it to play angry birds when no one is watching).
For me though (as someone who's an avid guitarist/pianist), the iPad is really rather useful, although a little limited at the moment. So, as an example, there's the iRig Amplitube app which turns your iPad into a guitar multiFX unit. On the iPad 1 it's a great little app, however there is a very slight signal delay compared to a Boss/Korg multi-FX unit, and it's also limited to being only able to run 4 FX at a time. I can *just* about cope with the delay if I wear headphones, but it's too much to deal with in a live situation imho. If the performance of the iPad3 is used to reduce that delay slightly, I'll be a very happy man (It turns out that running an iPad directly into a PA requires much less equipment than a multi-FX pedal, amp & cab!!)
So to answer your question, anyone running a lot of music apps on their tablets will notice the performance difference....
I was a lecturer at university, and at one point I did some sixth form teaching. I'd happily teach in a university again, but there is literally nothing in the world that would make me take a job in a school again. !! NOTHING !!
Those teachers deserve far more recognition for having to put up with the drivel that makes up the key-stages. I take my hat off to each and every one of them!
Yes. I taught plenty - some of whom I now work with. (I actually left teaching a number of years ago. Whilst the teaching itself was very rewarding, the government bureaucracy in the background just sucks your will to live).
The massive problem is the long term cost of decommissioning. I was at primary school when they started decommissioning my local nuclear plant. I'll be dead by the time they've finished..... That's one hell of a burden we are placing on our grand children.....
Do you like shadow boxing? you do? Oh good! Fight a patent troll in court, win, and the judge will order the troll to pay all legal expenses. And like that, poof. He's gone! (And your company is lumbered with all the legal costs). Sadly it's normally just cheaper to comply, no matter how much evidence of prior art you have.
A few years ago we used to have conferences where people would actually detail how the technical aspects of their games worked. These days a patent troll would be at the back of the conference, and would have submitted a patent application before the speaker had even finished presenting. It's no wonder that companies are so unwilling to let their employees write papers these days:(
Slashdot Mirror
Pentax doesn't have the clout (sadly). Hoya just sold them to Rioch.
It's still available on the UK site.....
Hint: 3G .....
Phone games are not exactly built to run on a specific hardware
Absolutely correct. No developer in their right mind would use any of the NEON vector instructions available on ARM CPUs. I mean come on, who could possibly want to process 4 floats in the time it takes to process 1? Especially when that code works on all of the ARM android phones too. I mean, what is the point in bothering to write scalable efficient code for a platform with a finite battery life? I can see absolutely no point to it at all. As a game developer myself, I can say with some authority that the entire iGaming industry has adopted the approach of running an SQL server on your iDevice, and then we simply fire an SQL query for every pixel when we come to rendering. It sure as hell beats using that crappy parallel computing language GLSL to render pixels....
Finally, there is device security. There has yet to be a single piece of malware on an iPad in the wild.
Apart from carrier IQ obviously....
can you do photoshop on an iPhone
Yes....
Admittedly it's not fully featured, however it's also worth bearing in mind that the iPhone/iPad are more powerful devices than the SGI machines I used in the late 90's to create CG animation. It is a matter of time until your phone is 'good enough' to do everything you need. For a large section of the population, I imagine that time has already arrived. For the rest of us, we'll just have to wait another 5 years or so...
One of the biggest myths in the industry, and one that is not exactly true. If you specifically select "Optimise for Intel i7", then guess what, it only runs on an i7. If however you specify the option to optimise for all platforms (/QxO) you'll have no such problems with AMD chips. The Intel compiler can be instructed to insert two codepaths into an exe. So you can have one specified as /QxO, and a separate codepath for i7. At startup, the app will indeed check for the intel vendor string, and will check to see if it's an i7. If it doesn't find it, it will run the codepath optimised for all processors, rather than the one specifically for the i7. In this regard there is absolutely no difference to running the exe on a Core2, or an AMD chip. In both cases the exe will run the second codepath. The problem only arises if the developer has simply chosen one codepath (or failed to select an optimised codepath for the second one, which could indeed look like a deliberate retardation of performance to a layman).
So the complaint about the CPU dispatcher basically boils down to this: Why does specifying "Optimise for Intel i7" mean exactly that, rather than "Optimise for any CPU supporting the same features as an i7". I can't answer that, and I'd certainly prefer the latter, however it is information that is clearly documented.
To be fair, you have a valid point about rigged benchmarks, but your reasoning is entirely incorrect. The simple fact is that the ATOM is significantly different enough from other x86 processors as to make any comparative benchmarking entirely redundant. You either compile a binary explicitly for ATOM (and it runs quite well), or you choose a general x86 compiler option (and it runs like a dog). Most benchmarking software will not have been compiled with the ATOM in mind, so chances are it will run like a dog, making the processor look worse than it actually is....
For anything that uses double-precision floating point the ARM core is very weak, although most of those things can be run on the DSP or GPU (and use even less power).
The ATOM core is extremely weak for double precision (a couple of orders of magnitude slower than single precision). I'm also quite surprised to hear that you've managed to get the GPU to support double precision? Which GPU are you using? The GL_DOUBLE is not a valid argument for glVertexAttribPointer in GLES, so I'd be interested in hearing how you subverted this? I've not tried windows 8 for tablets, so is this something microsoft have added to their D3D equivalent for tablets?
You are talking about the wrong department. This is G.C.H.Q. we are talking about, not Mi5 / Mi6. They are effectively nothing more than the poorly paid secretary to Mi5. Good programming graduates would go on to have interesting and fulfilling career in the software industry. Those that fail to do that, become teachers. Those that fail to do that, go to G.C.H.Q......
so you did it?
You seem to be assuming that people would want to work at GCHQ? I grew up next door to the place, and as someone who was training to be a programmer, GCHQ was that added bit of motivation to do well at university. It was always a case of "work hard, get a good degree, otherwise you'll have to apply to GCHQ.....". I certainly never got the impression it was an inspiring place to work.....
Or simply a high quality CG rendering (which technically speaking, won't be a photoshop job at all....)
I've heard those arguments said about the British when they first introduced recycling. Strangely enough, everyone recycles their waste just fine now. Maybe it was the education, maybe it was the councils refusal to collect mixed rubbish, or maybe it was the fines they started dishing out to people who constantly refused to recycle. Whatever it was, recycling and composting are second nature to the vast majority of the population now.... I don't see any reason why America should be any different.
I pay my city (I'm in the UK) to collect my recycling, and take away the very small quantity of stuff that's can't be recycled. The ONLY pre-requisite is that organic matter is not mixed in with the recycling. There is absolutely no need to sort the paper from glass.....
The problem of plastics is simply a case of "does it have the recycling symbol on it, or not". Even that seems to be moot these days (I'm assuming someone passed a law to say that all plastics containers must be able to be recycled, because I haven't seen a non-recyclable bit of plastic for years).
To be honest though, even with the councils previous "sort everything" system, I never found it to be too hard. Just collect your recycling in a single box, and sort it when you put it into the bins. It takes what, 2 minutes at most? Not exactly a high price to pay to keep it from being buried in the countryside....
Nothing. The life of the battery will stay the same, it'll just get bigger. Now it's just the job of marketing to persuade the consumer that retro styled 80's bricks are back in fashion.....
Will anyone ever see the performance advantage of a 4-core ARM chip outside of benchmarks?
I must admit I was very very sceptical of the iPad. I acquired a first gen one from the office fanboy, and I might have been somewhat converted to the platform. In almost all situations, I will admit that my cheapio netbook is much more practical tool (I've actually come to the conclusion that anyone using an iPad for 'business' is actually using it to play angry birds when no one is watching).
For me though (as someone who's an avid guitarist/pianist), the iPad is really rather useful, although a little limited at the moment. So, as an example, there's the iRig Amplitube app which turns your iPad into a guitar multiFX unit. On the iPad 1 it's a great little app, however there is a very slight signal delay compared to a Boss/Korg multi-FX unit, and it's also limited to being only able to run 4 FX at a time. I can *just* about cope with the delay if I wear headphones, but it's too much to deal with in a live situation imho. If the performance of the iPad3 is used to reduce that delay slightly, I'll be a very happy man (It turns out that running an iPad directly into a PA requires much less equipment than a multi-FX pedal, amp & cab!!)
So to answer your question, anyone running a lot of music apps on their tablets will notice the performance difference....
If it was based on the z80, I'd buy one.....
A slightly more informative article.
I was a lecturer at university, and at one point I did some sixth form teaching. I'd happily teach in a university again, but there is literally nothing in the world that would make me take a job in a school again. !! NOTHING !!
Those teachers deserve far more recognition for having to put up with the drivel that makes up the key-stages. I take my hat off to each and every one of them!
Yes. I taught plenty - some of whom I now work with. (I actually left teaching a number of years ago. Whilst the teaching itself was very rewarding, the government bureaucracy in the background just sucks your will to live).
No one is forcing you to drive a car.
Don't be so cruel to opiates! They relieve more pain and suffering than religion ever will.....
If true, this is good news. However, it could end up being bad news if this report gets twisted in support of the 'drill baby drill' crowd....
The massive problem is the long term cost of decommissioning. I was at primary school when they started decommissioning my local nuclear plant. I'll be dead by the time they've finished.....
That's one hell of a burden we are placing on our grand children.....
It would seem that business-ethics go on an extended sabbatical whenever a recession arrives.....
Do you like shadow boxing?
:(
you do?
Oh good!
Fight a patent troll in court, win, and the judge will order the troll to pay all legal expenses. And like that, poof. He's gone! (And your company is lumbered with all the legal costs). Sadly it's normally just cheaper to comply, no matter how much evidence of prior art you have.
A few years ago we used to have conferences where people would actually detail how the technical aspects of their games worked. These days a patent troll would be at the back of the conference, and would have submitted a patent application before the speaker had even finished presenting. It's no wonder that companies are so unwilling to let their employees write papers these days