14 replies to this topic

[Josh Klint]

Check out my performance tests of the iPad 2 and 3 with just under 10 million polygons:

[way2lazy2care]

Check out my performance tests of the iPad 2 and 3 with just under 10 million polygons:

"6-8fps... just about real time," bwah?

[Bacterius]

Ok but there's just a tiny problem with your demonstration. It's the same teapot repeated about four thousand times. Instancing is very misleading in terms of performance, perhaps having an actual scene drawn instead would be a better proof of concept...

[spooderw]

Ok but there's just a tiny problem with your demonstration. It's the same teapot repeated about four thousand times. Instancing is very misleading in terms of performance, perhaps having an actual scene drawn instead would be a better proof of concept...

I think the test was to see just how many polys could be rendered at once. Without instancing it usually becomes a bandwidth issue before it becomes a poly count issue.
So unless I misunderstood the point, instancing was an excellent way to measure the raw poly pushing power of the iPad.

[Bacterius]

I think the test was to see just how many polys could be rendered at once. Without instancing it usually becomes a bandwidth issue before it becomes a poly count issue.
So unless I misunderstood the point, instancing was an excellent way to measure the raw poly pushing power of the iPad.

It's a double-edged sword, as instanced polygons can be cached by the GPU which can then render them much, much faster than a for the same amount of polygons, even assuming infinite bandwidth.

[Bacterius]

Pretty cool nonetheless of course. I didn't know the iPad had that much horsepower. Sorry if I sounded blunt and derogatory, it wasn't my intention

[swiftcoder]

It's a double-edged sword, as instanced polygons can be cached by the GPU which can then render them much, much faster than a for the same amount of polygons, even assuming infinite bandwidth.

I think this might actually be a stickier point than you imagine.

For starters, OpenGL ES 2.0 has no explicit support for instancing. No ARB_draw_instanced, etc. So any instancing is of the old fashioned 'stick data in buffer and render variety'.

Given that we can only performa software instancing on iOS, does it make a huge difference whether we render 1 model or many? My gut feeling is no. In either case, all the vertex data should be resident in vertex buffers, so the only costs are binding buffers, and uploading modelview matrices. Matrix upload needs to occur in either case, so all we save is a few buffer binds - and given that one tends to pack multiple models into a single buffer on iOS, and that binds are not all that expensive... M guess would be that it doesn't make a huge difference.

[Promit]

Just like the iPhone 4(S), pushing polys themselves won't be much of a problem. The trouble is that the device is going to be so badly limited by pixel rate that you'll want to drag as much vertex pipeline as possible on to the CPU instead.

[Krohm]

Thank you for sharing Josh. Have you investigated batch length behavior?

[mdwh]

Just a comment on performance on the Ipad itself - I'm sure I remember hearing those kind of figures for GPUs *many* years ago, which is more than enough time for that to propagate down to smaller devices. E.g., people claim 90 million polygons per second the PowerVR SGX540 used in even small smartphones (e.g., http://smartphoneben...-arm-cortex-a9/ ). So I wouldn't be particularly surprised by this. Not to mention that millions of polygons isn't always a good indicator of performance.

I can't help feeling that this falls into the problem where netbooks are often viewed as "crap for games" because they can't run the latest AAA games, whilst a similarly sized tablet is viewed as amazing, because it can run 10 year old games...

Don't get me wrong, nothing wrong with your demo And the question of how best to optimise 3D code for the kind of GPUs (rather than the more usual AMD/NVIDIA/Intel combination) is indeed interesting and useful. I'm just not joining in the "wow, Ipads can do 3D" amazement. Unless of course one's initial expectation was that Ipads were completely crap ;)

[FableFox]

I can't help feeling that this falls into the problem where netbooks are often viewed as "crap for games" because they can't run the latest AAA games, whilst a similarly sized tablet is viewed as amazing, because it can run 10 year old games...

The keyword here is GPU, and architecture. I don't know the state of current netbook, but sadly, those days, netbook is seen what office worker carry when they doesn't want to break their back, specially just to do the usual office work (word, excel, etc). I don't remember any advertising point that talk about GPU and all.

Ipad, on the other hand, is "consume" device, and this, include multimedia and games. So it is tailored that way. The GPU on the new ipad is so powerful that it was 6 times (can't remember exact fact) than asus transformer or something - although the new ipad loses on the CPU side. Just to show the stress Apple put on graphic processing.

Sadly, when I browse Autodesk designer on itunes, it seems (from the top of my memory) it can only hold 6 layer. That is sad. I guess netbook has it uses and ipad 3 has it uses, and designed around that.

[zedz]

Not knocking your demo mate, but since when has anyone cared about the number of polygons
thats so last century (or whenever the geforce1 256 came out)
http://en.wikipedia.org/wiki/GeForce_256
hey it did come out last century, god Im getting on years

[meeshoo]

I think it would be useful if you could repeat the test with shaded polygons with common used shaders, from more simpler ones to more complex ones.

[JDX_John]

Just like the iPhone 4(S), pushing polys themselves won't be much of a problem. The trouble is that the device is going to be so badly limited by pixel rate that you'll want to drag as much vertex pipeline as possible on to the CPU instead.

On the Ogre forum someone reports the iPad3 is substantially slower than iPad2 for rendering at the device's native resolution, but substantially faster rendering at the iPad2's resolution of 1024x768... which would fit in nicely with claims the iPad3 has twice the GPU power while having to push 4X as many pixels.

[way2lazy2care]

I would also be curious about this.